{
  "$schema": "https://receiptaudit.com/benchmarks/schema.json",
  "suite": "receipt-audit",
  "title": "RA-Bench: receipt extraction, fake detection, evidence and duplicates",
  "preliminary": true,
  "disclosure": "We build Receipt Audit and we run this benchmark. Our engine is one of the targets, so read it as a vendor benchmark: the public split's cases, labels, every target's outputs and the code are open, every run is kept, and losses are listed first. To reproduce a number, run the command below with your own keys; to dispute one, email hello@godard.me.",
  "manifest": {
    "runId": "20261012T231905Z-2c62d5",
    "suite": "receipt-audit",
    "suiteVersion": "0.2.0",
    "datasetHash": "9cfa2c4c2d6e596579470768c8b182ffc7d264123e0c0fc7ee707a6de9dd6d79",
    "split": "private",
    "startedAt": "2026-10-12T23:19:05.681Z",
    "finishedAt": "2026-10-12T23:35:25.647Z",
    "gitSha": "3c7c53a7378b5e1704315156cbbef1ae521b89a5-dirty",
    "reps": 3,
    "cases": 151,
    "targets": [
      {
        "id": "self",
        "label": "Receipt Audit (Claude Haiku 5.5)",
        "kind": "self",
        "version": "engine.0.2.1+anthropic/claude-haiku-5.5+prompt.9a2a5877e979+git.3c7c53a7378b",
        "config": {
          "engine": "ReceiptChecks (apps/receipt-check/lib/server/ReceiptChecks.ts), in process",
          "engineVersion": "0.2.1",
          "production": true,
          "model": "anthropic/claude-haiku-5.5",
          "prompt": "apps/receipt-check/lib/receipts/model.ts (INSTRUCTIONS, USER_PROMPT and the ModelAnalysis output schema)",
          "promptSha256": "9a2a5877e9795c37df6d2c2cb89f56cae64b76ae10bbd672cc1ab6bb41426a71",
          "instructionsSha256": "da94bf411e9d3c47462a4d77b59a373a2b478a3e6afffb86889dc5e93ad452fb",
          "hashVersion": 2,
          "endpoint": "https://ai-gateway.vercel.sh/v1/chat/completions",
          "reasoningEffort": "low",
          "structuredOutput": true,
          "modelTimeout": "90 seconds",
          "modelRetries": "2, on errors effect/ai marks retryable",
          "filePreparation": "upright, at most 3072 px and 3.75 MB (lib/receipts/model.ts)",
          "duplicateNetwork": "the production Convex functions on an in-memory backend, empty at the start of each repetition",
          "submitter": "one account per case (the case id)",
          "gitSha": "3c7c53a7378b5e1704315156cbbef1ae521b89a5-dirty",
          "pricePerCheckUsd": 0.01633333333333333,
          "cost": "cost_per_check_usd is what a caller pays Receipt Audit per check: the list price of the smallest credit pack (credits-3k: $49 for 3,000 checks, $0.0163 per check). model_cost_per_check_usd (and each result's costUsd) is the engine's own model call: tokens × the AI Gateway list price on the price date."
        },
        "priceDate": "2026-10-12",
        "unsupportedTasks": []
      },
      {
        "id": "self-gemini",
        "label": "Receipt Audit on Gemini 3.8 Flash",
        "kind": "self",
        "version": "engine.0.2.1+google/gemini-3.8-flash+prompt.9a2a5877e979+git.3c7c53a7378b",
        "config": {
          "engine": "ReceiptChecks (apps/receipt-check/lib/server/ReceiptChecks.ts), in process",
          "engineVersion": "0.2.1",
          "production": false,
          "model": "google/gemini-3.8-flash",
          "prompt": "apps/receipt-check/lib/receipts/model.ts (INSTRUCTIONS, USER_PROMPT and the ModelAnalysis output schema)",
          "promptSha256": "9a2a5877e9795c37df6d2c2cb89f56cae64b76ae10bbd672cc1ab6bb41426a71",
          "instructionsSha256": "da94bf411e9d3c47462a4d77b59a373a2b478a3e6afffb86889dc5e93ad452fb",
          "hashVersion": 2,
          "endpoint": "https://ai-gateway.vercel.sh/v1/chat/completions",
          "reasoningEffort": "low",
          "structuredOutput": true,
          "modelTimeout": "90 seconds",
          "modelRetries": "2, on errors effect/ai marks retryable",
          "filePreparation": "upright, at most 3072 px and 3.75 MB (lib/receipts/model.ts)",
          "duplicateNetwork": "the production Convex functions on an in-memory backend, empty at the start of each repetition",
          "submitter": "one account per case (the case id)",
          "gitSha": "3c7c53a7378b5e1704315156cbbef1ae521b89a5-dirty",
          "pricePerCheckUsd": 0.01633333333333333,
          "cost": "cost_per_check_usd is what a caller pays Receipt Audit per check: the list price of the smallest credit pack (credits-3k: $49 for 3,000 checks, $0.0163 per check). model_cost_per_check_usd (and each result's costUsd) is the engine's own model call: tokens × the AI Gateway list price on the price date."
        },
        "priceDate": "2026-10-12",
        "unsupportedTasks": []
      },
      {
        "id": "diy:claude-haiku-5.5",
        "label": "Claude Haiku 5.5, plain prompt",
        "kind": "diy",
        "version": "claude-haiku-5-5@claude-code-cli+prompt.512dfb3893be+git.3c7c53a7378b",
        "config": {
          "transport": "claude-code-cli",
          "endpoint": "http://127.0.0.1:8789/v1/messages (claude-local: one headless `claude -p` per call, no tools, no MCP, no settings, empty cwd)",
          "billing": "subscription",
          "cli": "claude 2.1.296",
          "thinking": "Claude Code default (adaptive; can't be fully disabled via the CLI)",
          "ignored": [
            "max_tokens"
          ],
          "costUsd": "API-equivalent: the reported tokens × the AI Gateway list price, i.e. what the check costs when called through the API",
          "usage": "as the CLI reports it, including Claude Code's own cached system prompt (about 3,500 tokens, mostly cache reads); inputTokens counts uncached, cache-read and cache-write tokens",
          "model": "claude-haiku-5-5",
          "max_tokens": 4096,
          "message": "one user turn: the file (image block, or document block for a PDF), then the prompt",
          "rateLimits": "HTTP 429 (a subscription limit on claude-local) is retried with backoff like a transport error",
          "prompt": "apps/receipt-check/bench/prompt.ts (DIY_PLAIN_PROMPT)",
          "promptSha256": "512dfb3893be5aa7f89ff7be2259d8f9c94c948fe8b2d6e3dca9bc6f5889bc37",
          "gitSha": "3c7c53a7378b5e1704315156cbbef1ae521b89a5-dirty",
          "cost": "cost_per_check_usd and model_cost_per_check_usd are both the API-equivalent token cost: tokens × the AI Gateway list price on the price date"
        },
        "priceDate": "2026-10-12",
        "unsupportedTasks": [
          "checks",
          "duplicates",
          "proof"
        ]
      },
      {
        "id": "diy:gemini-3.8-flash",
        "label": "Gemini 3.8 Flash, plain prompt",
        "kind": "diy",
        "version": "google/gemini-3.8-flash+prompt.512dfb3893be+git.3c7c53a7378b",
        "config": {
          "endpoint": "https://ai-gateway.vercel.sh/v1/chat/completions (Vercel AI Gateway, OpenAI-compatible)",
          "model": "google/gemini-3.8-flash",
          "max_tokens": "not set (provider default)",
          "reasoning": "not set (provider default)",
          "temperature": "not set (provider default)",
          "message": "one user turn: the file (image_url part, or a file part for a PDF), then the prompt",
          "prompt": "apps/receipt-check/bench/prompt.ts (DIY_PLAIN_PROMPT)",
          "promptSha256": "512dfb3893be5aa7f89ff7be2259d8f9c94c948fe8b2d6e3dca9bc6f5889bc37",
          "gitSha": "3c7c53a7378b5e1704315156cbbef1ae521b89a5-dirty",
          "cost": "cost_per_check_usd and model_cost_per_check_usd are both the API-equivalent token cost: tokens × the AI Gateway list price on the price date"
        },
        "priceDate": "2026-10-12",
        "unsupportedTasks": [
          "checks",
          "duplicates",
          "proof"
        ]
      },
      {
        "id": "decision:gpt-6-luna",
        "label": "OpenAI decision model (gpt-6-luna), asked if it's fake",
        "kind": "diy",
        "version": "openai/gpt-6-luna-decisions+prompt.977e84771afc+git.3c7c53a7378b",
        "config": {
          "endpoint": "https://ai-gateway.vercel.sh/v1/decisions (Vercel AI Gateway → OpenAI Decisions API, public beta)",
          "model": "openai/gpt-6-luna-decisions",
          "questions": "apps/receipt-check/bench/prompt.ts (DECISION_QUESTIONS): a predicate (is it fake) and two choices (how it was made, which part looks altered)",
          "questionsSha256": "e17d0fb5b77b6147ffdb12fa1e22ed2dc99e5edec56d69731f297ad2c86aca37",
          "input": "one user message: DECISION_INPUT_TEXT and the image as a base64 data URL; a PDF's first page rendered at 150 dpi (pdftoppm)",
          "risk": "the predicate's probability",
          "reasons": "the two choices, in plain words",
          "cost": "input tokens × $0.1 per 1M (no output charge), price of 2026-10-12",
          "gitSha": "3c7c53a7378b5e1704315156cbbef1ae521b89a5-dirty"
        },
        "priceDate": "2026-10-12",
        "unsupportedTasks": [
          "extraction",
          "checks",
          "duplicates",
          "proof"
        ]
      }
    ]
  },
  "metrics": [
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "all",
      "target": "self",
      "value": 0.3298245614035083,
      "lo": 0.24351670349154178,
      "hi": 0.4293601063096113,
      "n": 95,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "all",
      "target": "self",
      "value": 0.3273809523809522,
      "lo": 0.24746070928115044,
      "hi": 0.4187497557776751,
      "n": 112,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "all",
      "target": "self",
      "value": 0.3333333333333329,
      "lo": 0.2436427243338542,
      "hi": 0.4369662978877758,
      "n": 88,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.subtotal",
      "stratum": "all",
      "target": "self",
      "value": 0.31249999999999994,
      "lo": 0.14164643854782033,
      "hi": 0.5559564416525933,
      "n": 16,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.tax",
      "stratum": "all",
      "target": "self",
      "value": 0.3333333333333332,
      "lo": 0.16906782991982014,
      "hi": 0.5513072227644528,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "all",
      "target": "self",
      "value": 0.3333333333333334,
      "lo": 0.255611110856911,
      "hi": 0.4213124551255819,
      "n": 121,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.receipt_number",
      "stratum": "all",
      "target": "self",
      "value": 0.33333333333333337,
      "lo": 0.132793283187926,
      "hi": 0.6201510391347513,
      "n": 11,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "all",
      "target": "self",
      "value": 0.32231404958677684,
      "lo": 0.24561669546616066,
      "hi": 0.4099464451267448,
      "n": 121,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "line_items_f1",
      "stratum": "all",
      "target": "self",
      "value": 0.4883720930232558,
      "lo": 0.463855421686747,
      "hi": 0.5,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "all",
      "target": "self",
      "value": 0.3333333333333334,
      "lo": 0.255611110856911,
      "hi": 0.4213124551255819,
      "n": 121,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "arithmetic=consistent",
      "target": "self",
      "value": 0.3288888888888886,
      "lo": 0.233188185723274,
      "hi": 0.44126397370417475,
      "n": 75,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "arithmetic=consistent",
      "target": "self",
      "value": 0.3260869565217387,
      "lo": 0.23894787729606762,
      "hi": 0.42716738854240865,
      "n": 92,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "arithmetic=consistent",
      "target": "self",
      "value": 0.3333333333333331,
      "lo": 0.23593569195030484,
      "hi": 0.4473949773682717,
      "n": 73,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.subtotal",
      "stratum": "arithmetic=consistent",
      "target": "self",
      "value": 0.30303030303030304,
      "lo": 0.11471049994616331,
      "hi": 0.5933146082533644,
      "n": 11,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.tax",
      "stratum": "arithmetic=consistent",
      "target": "self",
      "value": 0.3333333333333333,
      "lo": 0.1430225574285459,
      "hi": 0.5996759000228535,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "arithmetic=consistent",
      "target": "self",
      "value": 0.3333333333333329,
      "lo": 0.24862102140164288,
      "hi": 0.43037681133401817,
      "n": 100,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "arithmetic=consistent",
      "target": "self",
      "value": 0.31999999999999956,
      "lo": 0.2366914732499905,
      "hi": 0.4166261861045234,
      "n": 100,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "line_items_f1",
      "stratum": "arithmetic=consistent",
      "target": "self",
      "value": 0.48214285714285715,
      "lo": 0.44339622641509435,
      "hi": 0.5,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "arithmetic=consistent",
      "target": "self",
      "value": 0.3333333333333329,
      "lo": 0.24862102140164288,
      "hi": 0.43037681133401817,
      "n": 100,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "arithmetic=inconsistent",
      "target": "self",
      "value": 0.33333333333333326,
      "lo": 0.15935010395826088,
      "hi": 0.5687559410446896,
      "n": 17,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "arithmetic=inconsistent",
      "target": "self",
      "value": 0.33333333333333326,
      "lo": 0.1627877475083369,
      "hi": 0.5625053270405489,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "arithmetic=inconsistent",
      "target": "self",
      "value": 0.3333333333333333,
      "lo": 0.1475552363427626,
      "hi": 0.5908816979031161,
      "n": 14,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "arithmetic=inconsistent",
      "target": "self",
      "value": 0.33333333333333326,
      "lo": 0.1627877475083369,
      "hi": 0.5625053270405489,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "arithmetic=inconsistent",
      "target": "self",
      "value": 0.33333333333333326,
      "lo": 0.1627877475083369,
      "hi": 0.5625053270405489,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "arithmetic=inconsistent",
      "target": "self",
      "value": 0.33333333333333326,
      "lo": 0.1627877475083369,
      "hi": 0.5625053270405489,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "base=sroie",
      "target": "self",
      "value": 0.33333333333333326,
      "lo": 0.15935010395826088,
      "hi": 0.5687559410446896,
      "n": 17,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "base=sroie",
      "target": "self",
      "value": 0.3333333333333332,
      "lo": 0.16906782991982014,
      "hi": 0.5513072227644528,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "base=sroie",
      "target": "self",
      "value": 0.33333333333333326,
      "lo": 0.15568427411889454,
      "hi": 0.5755182860592508,
      "n": 16,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "base=sroie",
      "target": "self",
      "value": 0.3333333333333332,
      "lo": 0.16906782991982014,
      "hi": 0.5513072227644528,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "base=sroie",
      "target": "self",
      "value": 0.3333333333333332,
      "lo": 0.16906782991982014,
      "hi": 0.5513072227644528,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "base=sroie",
      "target": "self",
      "value": 0.3333333333333332,
      "lo": 0.16906782991982014,
      "hi": 0.5513072227644528,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "capture=photo",
      "target": "self",
      "value": 0.3333333333333333,
      "lo": 0.1475552363427626,
      "hi": 0.5908816979031161,
      "n": 14,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "capture=photo",
      "target": "self",
      "value": 0.3333333333333332,
      "lo": 0.17194752605083474,
      "hi": 0.5462654802574475,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "capture=photo",
      "target": "self",
      "value": 0.3333333333333333,
      "lo": 0.1430225574285459,
      "hi": 0.5996759000228535,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.tax",
      "stratum": "capture=photo",
      "target": "self",
      "value": 0.3333333333333333,
      "lo": 0.13812009109121315,
      "hi": 0.6093779111272004,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "capture=photo",
      "target": "self",
      "value": 0.3333333333333332,
      "lo": 0.17194752605083474,
      "hi": 0.5462654802574475,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "capture=photo",
      "target": "self",
      "value": 0.3333333333333332,
      "lo": 0.17194752605083474,
      "hi": 0.5462654802574475,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "line_items_f1",
      "stratum": "capture=photo",
      "target": "self",
      "value": 0.4807692307692308,
      "lo": 0.4387755102040816,
      "hi": 0.5,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "capture=photo",
      "target": "self",
      "value": 0.3333333333333332,
      "lo": 0.17194752605083474,
      "hi": 0.5462654802574475,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "capture=scan",
      "target": "self",
      "value": 0.32921810699588444,
      "lo": 0.2366576417318615,
      "hi": 0.4372439265434072,
      "n": 81,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "capture=scan",
      "target": "self",
      "value": 0.3259259259259255,
      "lo": 0.23795043525097515,
      "hi": 0.42815307883438225,
      "n": 90,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "capture=scan",
      "target": "self",
      "value": 0.33333333333333304,
      "lo": 0.236514174104019,
      "hi": 0.44660241902810915,
      "n": 74,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "capture=scan",
      "target": "self",
      "value": 0.3333333333333329,
      "lo": 0.24823816410787972,
      "hi": 0.43087957324806203,
      "n": 99,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "capture=scan",
      "target": "self",
      "value": 0.3198653198653194,
      "lo": 0.2361974610245463,
      "hi": 0.4169903965217745,
      "n": 99,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "capture=scan",
      "target": "self",
      "value": 0.3333333333333329,
      "lo": 0.24823816410787972,
      "hi": 0.43087957324806203,
      "n": 99,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "class=ai_full",
      "target": "self",
      "value": 0.3333333333333332,
      "lo": 0.16602033991933535,
      "hi": 0.5567060680982032,
      "n": 19,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "class=ai_full",
      "target": "self",
      "value": 0.3333333333333332,
      "lo": 0.16906782991982014,
      "hi": 0.5513072227644528,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "class=ai_full",
      "target": "self",
      "value": 0.3333333333333333,
      "lo": 0.13812009109121315,
      "hi": 0.6093779111272004,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.subtotal",
      "stratum": "class=ai_full",
      "target": "self",
      "value": 0.31249999999999994,
      "lo": 0.14164643854782033,
      "hi": 0.5559564416525933,
      "n": 16,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.tax",
      "stratum": "class=ai_full",
      "target": "self",
      "value": 0.3333333333333332,
      "lo": 0.16906782991982014,
      "hi": 0.5513072227644528,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "class=ai_full",
      "target": "self",
      "value": 0.3333333333333332,
      "lo": 0.16906782991982014,
      "hi": 0.5513072227644528,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.receipt_number",
      "stratum": "class=ai_full",
      "target": "self",
      "value": 0.33333333333333337,
      "lo": 0.132793283187926,
      "hi": 0.6201510391347513,
      "n": 11,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "class=ai_full",
      "target": "self",
      "value": 0.31666666666666654,
      "lo": 0.15716056329227918,
      "hi": 0.5352519946604211,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "line_items_f1",
      "stratum": "class=ai_full",
      "target": "self",
      "value": 0.4883720930232558,
      "lo": 0.46296296296296297,
      "hi": 0.5,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "class=ai_full",
      "target": "self",
      "value": 0.3333333333333332,
      "lo": 0.16906782991982014,
      "hi": 0.5513072227644528,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "class=ai_inpaint",
      "target": "self",
      "value": 0.33333333333333337,
      "lo": 0.1269754121301785,
      "hi": 0.632202187824118,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "class=ai_inpaint",
      "target": "self",
      "value": 0.33333333333333337,
      "lo": 0.1269754121301785,
      "hi": 0.632202187824118,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "class=ai_inpaint",
      "target": "self",
      "value": 0.33333333333333337,
      "lo": 0.1269754121301785,
      "hi": 0.632202187824118,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "class=ai_inpaint",
      "target": "self",
      "value": 0.33333333333333337,
      "lo": 0.1269754121301785,
      "hi": 0.632202187824118,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "class=ai_inpaint",
      "target": "self",
      "value": 0.33333333333333337,
      "lo": 0.1269754121301785,
      "hi": 0.632202187824118,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "class=ai_inpaint",
      "target": "self",
      "value": 0.33333333333333337,
      "lo": 0.1269754121301785,
      "hi": 0.632202187824118,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "class=genuine",
      "target": "self",
      "value": 0.32653061224489816,
      "lo": 0.21207497001811879,
      "hi": 0.46620794685429856,
      "n": 49,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "class=genuine",
      "target": "self",
      "value": 0.32258064516129026,
      "lo": 0.21954490651283887,
      "hi": 0.446319119071519,
      "n": 62,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "class=genuine",
      "target": "self",
      "value": 0.3333333333333335,
      "lo": 0.21969931853588529,
      "hi": 0.47031622143029206,
      "n": 51,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "class=genuine",
      "target": "self",
      "value": 0.3333333333333331,
      "lo": 0.23413420812686653,
      "hi": 0.449873479343222,
      "n": 70,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "class=genuine",
      "target": "self",
      "value": 0.31904761904761886,
      "lo": 0.22173291118173366,
      "hi": 0.43518972092864827,
      "n": 70,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "class=genuine",
      "target": "self",
      "value": 0.3333333333333331,
      "lo": 0.23413420812686653,
      "hi": 0.449873479343222,
      "n": 70,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "class=hard_negative",
      "target": "self",
      "value": 0.33333333333333337,
      "lo": 0.1269754121301785,
      "hi": 0.632202187824118,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "class=hard_negative",
      "target": "self",
      "value": 0.33333333333333337,
      "lo": 0.1269754121301785,
      "hi": 0.632202187824118,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "class=hard_negative",
      "target": "self",
      "value": 0.33333333333333337,
      "lo": 0.132793283187926,
      "hi": 0.6201510391347513,
      "n": 11,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "class=hard_negative",
      "target": "self",
      "value": 0.33333333333333337,
      "lo": 0.132793283187926,
      "hi": 0.6201510391347513,
      "n": 11,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "class=hard_negative",
      "target": "self",
      "value": 0.33333333333333337,
      "lo": 0.132793283187926,
      "hi": 0.6201510391347513,
      "n": 11,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "class=manual_edit",
      "target": "self",
      "value": 0.33333333333333337,
      "lo": 0.1269754121301785,
      "hi": 0.632202187824118,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "class=manual_edit",
      "target": "self",
      "value": 0.33333333333333337,
      "lo": 0.1269754121301785,
      "hi": 0.632202187824118,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "class=manual_edit",
      "target": "self",
      "value": 0.33333333333333337,
      "lo": 0.1269754121301785,
      "hi": 0.632202187824118,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "class=manual_edit",
      "target": "self",
      "value": 0.33333333333333337,
      "lo": 0.1269754121301785,
      "hi": 0.632202187824118,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "country=MY",
      "target": "self",
      "value": 0.3290043290043287,
      "lo": 0.23438633990053026,
      "hi": 0.43987320813978203,
      "n": 77,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "country=MY",
      "target": "self",
      "value": 0.325396825396825,
      "lo": 0.23475730445190776,
      "hi": 0.4313077435659093,
      "n": 84,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "country=MY",
      "target": "self",
      "value": 0.3333333333333331,
      "lo": 0.2347460591364293,
      "hi": 0.44902992492920035,
      "n": 71,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "country=MY",
      "target": "self",
      "value": 0.33333333333333287,
      "lo": 0.24582386759676284,
      "hi": 0.43406529994048143,
      "n": 93,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "country=MY",
      "target": "self",
      "value": 0.3189964157706089,
      "lo": 0.2330698296477433,
      "hi": 0.4192829221937801,
      "n": 93,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "country=MY",
      "target": "self",
      "value": 0.33333333333333287,
      "lo": 0.24582386759676284,
      "hi": 0.43406529994048143,
      "n": 93,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "currency=MYR",
      "target": "self",
      "value": 0.3290043290043287,
      "lo": 0.23438633990053026,
      "hi": 0.43987320813978203,
      "n": 77,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "currency=MYR",
      "target": "self",
      "value": 0.325396825396825,
      "lo": 0.23475730445190776,
      "hi": 0.4313077435659093,
      "n": 84,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "currency=MYR",
      "target": "self",
      "value": 0.3333333333333331,
      "lo": 0.2347460591364293,
      "hi": 0.44902992492920035,
      "n": 71,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "currency=MYR",
      "target": "self",
      "value": 0.33333333333333287,
      "lo": 0.24582386759676284,
      "hi": 0.43406529994048143,
      "n": 93,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "currency=MYR",
      "target": "self",
      "value": 0.3189964157706089,
      "lo": 0.2330698296477433,
      "hi": 0.4192829221937801,
      "n": 93,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "currency=MYR",
      "target": "self",
      "value": 0.33333333333333287,
      "lo": 0.24582386759676284,
      "hi": 0.43406529994048143,
      "n": 93,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "dateEra=old",
      "target": "self",
      "value": 0.32926829268292646,
      "lo": 0.23720116686677414,
      "hi": 0.4366161192713083,
      "n": 82,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "dateEra=old",
      "target": "self",
      "value": 0.32584269662921306,
      "lo": 0.23743972450293455,
      "hi": 0.42865772383778183,
      "n": 89,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "dateEra=old",
      "target": "self",
      "value": 0.33333333333333304,
      "lo": 0.236514174104019,
      "hi": 0.44660241902810915,
      "n": 74,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.tax",
      "stratum": "dateEra=old",
      "target": "self",
      "value": 0.33333333333333337,
      "lo": 0.1269754121301785,
      "hi": 0.632202187824118,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "dateEra=old",
      "target": "self",
      "value": 0.33333333333333287,
      "lo": 0.24784994777000835,
      "hi": 0.43139004893721755,
      "n": 98,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "dateEra=old",
      "target": "self",
      "value": 0.31972789115646216,
      "lo": 0.23569598485015364,
      "hi": 0.41735952179235625,
      "n": 98,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "line_items_f1",
      "stratum": "dateEra=old",
      "target": "self",
      "value": 0.5,
      "lo": 0.5,
      "hi": 0.5,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "dateEra=old",
      "target": "self",
      "value": 0.33333333333333287,
      "lo": 0.24784994777000835,
      "hi": 0.43139004893721755,
      "n": 98,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "dateEra=recent",
      "target": "self",
      "value": 0.3333333333333333,
      "lo": 0.1430225574285459,
      "hi": 0.5996759000228535,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "dateEra=recent",
      "target": "self",
      "value": 0.3333333333333332,
      "lo": 0.17726209796920395,
      "hi": 0.5371101081809382,
      "n": 23,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "dateEra=recent",
      "target": "self",
      "value": 0.3333333333333333,
      "lo": 0.1475552363427626,
      "hi": 0.5908816979031161,
      "n": 14,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.tax",
      "stratum": "dateEra=recent",
      "target": "self",
      "value": 0.33333333333333337,
      "lo": 0.1269754121301785,
      "hi": 0.632202187824118,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "dateEra=recent",
      "target": "self",
      "value": 0.3333333333333332,
      "lo": 0.17726209796920395,
      "hi": 0.5371101081809382,
      "n": 23,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "dateEra=recent",
      "target": "self",
      "value": 0.3333333333333332,
      "lo": 0.17726209796920395,
      "hi": 0.5371101081809382,
      "n": 23,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "line_items_f1",
      "stratum": "dateEra=recent",
      "target": "self",
      "value": 0.4772727272727273,
      "lo": 0.4268292682926829,
      "hi": 0.5,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "dateEra=recent",
      "target": "self",
      "value": 0.3333333333333332,
      "lo": 0.17726209796920395,
      "hi": 0.5371101081809382,
      "n": 23,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "docType=receipt",
      "target": "self",
      "value": 0.3298245614035083,
      "lo": 0.24351670349154178,
      "hi": 0.4293601063096113,
      "n": 95,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "docType=receipt",
      "target": "self",
      "value": 0.3273809523809522,
      "lo": 0.24746070928115044,
      "hi": 0.4187497557776751,
      "n": 112,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "docType=receipt",
      "target": "self",
      "value": 0.3333333333333329,
      "lo": 0.2436427243338542,
      "hi": 0.4369662978877758,
      "n": 88,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.subtotal",
      "stratum": "docType=receipt",
      "target": "self",
      "value": 0.31249999999999994,
      "lo": 0.14164643854782033,
      "hi": 0.5559564416525933,
      "n": 16,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.tax",
      "stratum": "docType=receipt",
      "target": "self",
      "value": 0.3333333333333332,
      "lo": 0.16906782991982014,
      "hi": 0.5513072227644528,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "docType=receipt",
      "target": "self",
      "value": 0.3333333333333334,
      "lo": 0.255611110856911,
      "hi": 0.4213124551255819,
      "n": 121,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.receipt_number",
      "stratum": "docType=receipt",
      "target": "self",
      "value": 0.33333333333333337,
      "lo": 0.132793283187926,
      "hi": 0.6201510391347513,
      "n": 11,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "docType=receipt",
      "target": "self",
      "value": 0.32231404958677684,
      "lo": 0.24561669546616066,
      "hi": 0.4099464451267448,
      "n": 121,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "line_items_f1",
      "stratum": "docType=receipt",
      "target": "self",
      "value": 0.4883720930232558,
      "lo": 0.463855421686747,
      "hi": 0.5,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "docType=receipt",
      "target": "self",
      "value": 0.3333333333333334,
      "lo": 0.255611110856911,
      "hi": 0.4213124551255819,
      "n": 121,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "lang=en",
      "target": "self",
      "value": 0.3293172690763049,
      "lo": 0.23773552188620384,
      "hi": 0.43599942317230067,
      "n": 83,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "lang=en",
      "target": "self",
      "value": 0.3259259259259255,
      "lo": 0.23795043525097515,
      "hi": 0.42815307883438225,
      "n": 90,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "lang=en",
      "target": "self",
      "value": 0.33333333333333304,
      "lo": 0.236514174104019,
      "hi": 0.44660241902810915,
      "n": 74,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.tax",
      "stratum": "lang=en",
      "target": "self",
      "value": 0.33333333333333337,
      "lo": 0.1269754121301785,
      "hi": 0.632202187824118,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "lang=en",
      "target": "self",
      "value": 0.3333333333333329,
      "lo": 0.24823816410787972,
      "hi": 0.43087957324806203,
      "n": 99,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "lang=en",
      "target": "self",
      "value": 0.3198653198653194,
      "lo": 0.2361974610245463,
      "hi": 0.4169903965217745,
      "n": 99,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "line_items_f1",
      "stratum": "lang=en",
      "target": "self",
      "value": 0.47560975609756095,
      "lo": 0.425,
      "hi": 0.5,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "lang=en",
      "target": "self",
      "value": 0.3333333333333329,
      "lo": 0.24823816410787972,
      "hi": 0.43087957324806203,
      "n": 99,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "provenance=intact",
      "target": "self",
      "value": 0.33333333333333326,
      "lo": 0.15935010395826088,
      "hi": 0.5687559410446896,
      "n": 17,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "provenance=intact",
      "target": "self",
      "value": 0.3333333333333332,
      "lo": 0.16906782991982014,
      "hi": 0.5513072227644528,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "provenance=intact",
      "target": "self",
      "value": 0.3333333333333333,
      "lo": 0.13812009109121315,
      "hi": 0.6093779111272004,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.tax",
      "stratum": "provenance=intact",
      "target": "self",
      "value": 0.33333333333333337,
      "lo": 0.1269754121301785,
      "hi": 0.632202187824118,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "provenance=intact",
      "target": "self",
      "value": 0.3333333333333332,
      "lo": 0.16906782991982014,
      "hi": 0.5513072227644528,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "provenance=intact",
      "target": "self",
      "value": 0.3333333333333332,
      "lo": 0.16906782991982014,
      "hi": 0.5513072227644528,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "line_items_f1",
      "stratum": "provenance=intact",
      "target": "self",
      "value": 0.5,
      "lo": 0.5,
      "hi": 0.5,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "provenance=intact",
      "target": "self",
      "value": 0.3333333333333332,
      "lo": 0.16906782991982014,
      "hi": 0.5513072227644528,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "provenance=none",
      "target": "self",
      "value": 0.327683615819209,
      "lo": 0.22169172275612037,
      "hi": 0.4547426930751457,
      "n": 59,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "provenance=none",
      "target": "self",
      "value": 0.32407407407407385,
      "lo": 0.2272752025733733,
      "hi": 0.43869465445130584,
      "n": 72,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "provenance=none",
      "target": "self",
      "value": 0.33333333333333326,
      "lo": 0.22729261523522676,
      "hi": 0.45943133547880305,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "provenance=none",
      "target": "self",
      "value": 0.333333333333333,
      "lo": 0.2402877124166734,
      "hi": 0.4414716493941309,
      "n": 81,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "provenance=none",
      "target": "self",
      "value": 0.32098765432098736,
      "lo": 0.22942221969038784,
      "hi": 0.4287637615138095,
      "n": 81,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "provenance=none",
      "target": "self",
      "value": 0.333333333333333,
      "lo": 0.2402877124166734,
      "hi": 0.4414716493941309,
      "n": 81,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "provenance=stripped",
      "target": "self",
      "value": 0.3333333333333332,
      "lo": 0.16602033991933535,
      "hi": 0.5567060680982032,
      "n": 19,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "provenance=stripped",
      "target": "self",
      "value": 0.3333333333333332,
      "lo": 0.16906782991982014,
      "hi": 0.5513072227644528,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "provenance=stripped",
      "target": "self",
      "value": 0.33333333333333326,
      "lo": 0.15568427411889454,
      "hi": 0.5755182860592508,
      "n": 16,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.tax",
      "stratum": "provenance=stripped",
      "target": "self",
      "value": 0.33333333333333337,
      "lo": 0.1269754121301785,
      "hi": 0.632202187824118,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "provenance=stripped",
      "target": "self",
      "value": 0.3333333333333332,
      "lo": 0.16906782991982014,
      "hi": 0.5513072227644528,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "provenance=stripped",
      "target": "self",
      "value": 0.31666666666666654,
      "lo": 0.15716056329227918,
      "hi": 0.5352519946604211,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "line_items_f1",
      "stratum": "provenance=stripped",
      "target": "self",
      "value": 0.47674418604651164,
      "lo": 0.425,
      "hi": 0.5,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "provenance=stripped",
      "target": "self",
      "value": 0.3333333333333332,
      "lo": 0.16906782991982014,
      "hi": 0.5513072227644528,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "source=commons",
      "target": "self",
      "value": 0.3333333333333333,
      "lo": 0.1430225574285459,
      "hi": 0.5996759000228535,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "source=commons",
      "target": "self",
      "value": 0.3333333333333333,
      "lo": 0.1430225574285459,
      "hi": 0.5996759000228535,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "source=commons",
      "target": "self",
      "value": 0.3333333333333333,
      "lo": 0.1430225574285459,
      "hi": 0.5996759000228535,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "source=commons",
      "target": "self",
      "value": 0.3333333333333333,
      "lo": 0.1430225574285459,
      "hi": 0.5996759000228535,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "source=edited:sharp-patch",
      "target": "self",
      "value": 0.33333333333333337,
      "lo": 0.1269754121301785,
      "hi": 0.632202187824118,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "source=edited:sharp-patch",
      "target": "self",
      "value": 0.33333333333333337,
      "lo": 0.1269754121301785,
      "hi": 0.632202187824118,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "source=edited:sharp-patch",
      "target": "self",
      "value": 0.33333333333333337,
      "lo": 0.1269754121301785,
      "hi": 0.632202187824118,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "source=edited:sharp-patch",
      "target": "self",
      "value": 0.33333333333333337,
      "lo": 0.1269754121301785,
      "hi": 0.632202187824118,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "source=sroie",
      "target": "self",
      "value": 0.3272727272727273,
      "lo": 0.21813065595168943,
      "hi": 0.45896776443421317,
      "n": 55,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "source=sroie",
      "target": "self",
      "value": 0.3220338983050847,
      "lo": 0.21687575241885632,
      "hi": 0.44894995573474633,
      "n": 59,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "source=sroie",
      "target": "self",
      "value": 0.3333333333333334,
      "lo": 0.22062697673438705,
      "hi": 0.46897043530690063,
      "n": 52,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "source=sroie",
      "target": "self",
      "value": 0.33333333333333315,
      "lo": 0.23287456398365092,
      "hi": 0.451615881523039,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "source=sroie",
      "target": "self",
      "value": 0.318627450980392,
      "lo": 0.22014575625387006,
      "hi": 0.436505610915175,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "source=sroie",
      "target": "self",
      "value": 0.33333333333333315,
      "lo": 0.23287456398365092,
      "hi": 0.451615881523039,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "totals_add_up.recall",
      "stratum": "all",
      "target": "self",
      "value": 0.12962962962962962,
      "lo": 0.03957024141337484,
      "hi": 0.34996992425081597,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "totals_add_up.precision",
      "stratum": "all",
      "target": "self",
      "value": 0.8749999999999999,
      "lo": 0.31667744818951943,
      "hi": 0.9906307012744284,
      "n": 8,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "false_alarm_rate",
      "stratum": "all",
      "target": "self",
      "value": 0.004273504273504273,
      "lo": 0.0003178190465284274,
      "hi": 0.054765821351886415,
      "n": 78,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "checks",
      "metric": "coverage",
      "stratum": "all",
      "target": "self",
      "value": 0.1983471074380164,
      "lo": 0.13706808314327465,
      "hi": 0.278190271816774,
      "n": 121,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "false_alarm_rate",
      "stratum": "arithmetic=consistent",
      "target": "self",
      "value": 0.004273504273504273,
      "lo": 0.0003178190465284274,
      "hi": 0.054765821351886415,
      "n": 78,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "checks",
      "metric": "coverage",
      "stratum": "arithmetic=consistent",
      "target": "self",
      "value": 0.20999999999999994,
      "lo": 0.14165654061915295,
      "hi": 0.2997996883408986,
      "n": 100,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "totals_add_up.recall",
      "stratum": "arithmetic=inconsistent",
      "target": "self",
      "value": 0.12962962962962962,
      "lo": 0.03957024141337484,
      "hi": 0.34996992425081597,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "coverage",
      "stratum": "arithmetic=inconsistent",
      "target": "self",
      "value": 0.12962962962962962,
      "lo": 0.03957024141337484,
      "hi": 0.34996992425081597,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "totals_add_up.recall",
      "stratum": "base=sroie",
      "target": "self",
      "value": 0.1818181818181818,
      "lo": 0.05136768974608508,
      "hi": 0.47698056196084415,
      "n": 11,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "coverage",
      "stratum": "base=sroie",
      "target": "self",
      "value": 0.16666666666666669,
      "lo": 0.06143139945263601,
      "hi": 0.3793187059159103,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "coverage",
      "stratum": "capture=photo",
      "target": "self",
      "value": 0.19047619047619047,
      "lo": 0.07667565052761793,
      "hi": 0.40000564690204926,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "totals_add_up.recall",
      "stratum": "capture=scan",
      "target": "self",
      "value": 0.17948717948717946,
      "lo": 0.05527092607546122,
      "hi": 0.44991841517722997,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "false_alarm_rate",
      "stratum": "capture=scan",
      "target": "self",
      "value": 0.004901960784313725,
      "lo": 0.0003645718312947112,
      "hi": 0.06238645746799082,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "checks",
      "metric": "coverage",
      "stratum": "capture=scan",
      "target": "self",
      "value": 0.202020202020202,
      "lo": 0.13475729804135694,
      "hi": 0.2915441111707826,
      "n": 99,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "coverage",
      "stratum": "class=ai_full",
      "target": "self",
      "value": 0.11666666666666665,
      "lo": 0.035533746172033476,
      "hi": 0.3213288750017947,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "coverage",
      "stratum": "class=ai_inpaint",
      "target": "self",
      "value": 0.1,
      "lo": 0.017876213095072896,
      "hi": 0.4041500267952385,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "false_alarm_rate",
      "stratum": "class=genuine",
      "target": "self",
      "value": 0.004975124378109453,
      "lo": 0.0003700149377350512,
      "hi": 0.06326680963375278,
      "n": 67,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "checks",
      "metric": "coverage",
      "stratum": "class=genuine",
      "target": "self",
      "value": 0.23333333333333348,
      "lo": 0.14974434785455806,
      "hi": 0.3446679520975845,
      "n": 70,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "false_alarm_rate",
      "stratum": "class=hard_negative",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.2588329669680317,
      "n": 11,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "checks",
      "metric": "coverage",
      "stratum": "class=hard_negative",
      "target": "self",
      "value": 0.1818181818181818,
      "lo": 0.05136768974608508,
      "hi": 0.47698056196084415,
      "n": 11,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "coverage",
      "stratum": "class=manual_edit",
      "target": "self",
      "value": 0.2333333333333333,
      "lo": 0.07255530696407597,
      "hi": 0.5421288529627982,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "totals_add_up.recall",
      "stratum": "country=MY",
      "target": "self",
      "value": 0.17948717948717946,
      "lo": 0.05527092607546122,
      "hi": 0.44991841517722997,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "false_alarm_rate",
      "stratum": "country=MY",
      "target": "self",
      "value": 0.00505050505050505,
      "lo": 0.0003756230402271882,
      "hi": 0.06417236269034904,
      "n": 66,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "checks",
      "metric": "coverage",
      "stratum": "country=MY",
      "target": "self",
      "value": 0.2150537634408602,
      "lo": 0.1437502200694047,
      "hi": 0.30896351797814325,
      "n": 93,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "totals_add_up.recall",
      "stratum": "currency=MYR",
      "target": "self",
      "value": 0.17948717948717946,
      "lo": 0.05527092607546122,
      "hi": 0.44991841517722997,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "false_alarm_rate",
      "stratum": "currency=MYR",
      "target": "self",
      "value": 0.00505050505050505,
      "lo": 0.0003756230402271882,
      "hi": 0.06417236269034904,
      "n": 66,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "checks",
      "metric": "coverage",
      "stratum": "currency=MYR",
      "target": "self",
      "value": 0.2150537634408602,
      "lo": 0.1437502200694047,
      "hi": 0.30896351797814325,
      "n": 93,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "totals_add_up.recall",
      "stratum": "dateEra=old",
      "target": "self",
      "value": 0.16666666666666666,
      "lo": 0.0512064627237552,
      "hi": 0.425667405768002,
      "n": 14,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "false_alarm_rate",
      "stratum": "dateEra=old",
      "target": "self",
      "value": 0.00505050505050505,
      "lo": 0.0003756230402271882,
      "hi": 0.06417236269034904,
      "n": 66,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "checks",
      "metric": "coverage",
      "stratum": "dateEra=old",
      "target": "self",
      "value": 0.20748299319727886,
      "lo": 0.13899198894461945,
      "hi": 0.2980414746639827,
      "n": 98,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "false_alarm_rate",
      "stratum": "dateEra=recent",
      "target": "self",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "checks",
      "metric": "coverage",
      "stratum": "dateEra=recent",
      "target": "self",
      "value": 0.15942028985507248,
      "lo": 0.06134924515547946,
      "hi": 0.3549765674122026,
      "n": 23,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "totals_add_up.recall",
      "stratum": "docType=receipt",
      "target": "self",
      "value": 0.12962962962962962,
      "lo": 0.03957024141337484,
      "hi": 0.34996992425081597,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "false_alarm_rate",
      "stratum": "docType=receipt",
      "target": "self",
      "value": 0.004273504273504273,
      "lo": 0.0003178190465284274,
      "hi": 0.054765821351886415,
      "n": 78,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "checks",
      "metric": "coverage",
      "stratum": "docType=receipt",
      "target": "self",
      "value": 0.1983471074380164,
      "lo": 0.13706808314327465,
      "hi": 0.278190271816774,
      "n": 121,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "totals_add_up.recall",
      "stratum": "lang=en",
      "target": "self",
      "value": 0.16666666666666666,
      "lo": 0.0512064627237552,
      "hi": 0.425667405768002,
      "n": 14,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "false_alarm_rate",
      "stratum": "lang=en",
      "target": "self",
      "value": 0.004975124378109453,
      "lo": 0.0003700149377350512,
      "hi": 0.06326680963375278,
      "n": 67,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "checks",
      "metric": "coverage",
      "stratum": "lang=en",
      "target": "self",
      "value": 0.2087542087542087,
      "lo": 0.14033499103370675,
      "hi": 0.2989313580832431,
      "n": 99,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "coverage",
      "stratum": "provenance=intact",
      "target": "self",
      "value": 0.13333333333333333,
      "lo": 0.043720722486203234,
      "hi": 0.3411043934191976,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "false_alarm_rate",
      "stratum": "provenance=none",
      "target": "self",
      "value": 0.004273504273504273,
      "lo": 0.0003178190465284274,
      "hi": 0.054765821351886415,
      "n": 78,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "checks",
      "metric": "coverage",
      "stratum": "provenance=none",
      "target": "self",
      "value": 0.2263374485596708,
      "lo": 0.14882771770067627,
      "hi": 0.32862901218620116,
      "n": 81,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "totals_add_up.recall",
      "stratum": "provenance=stripped",
      "target": "self",
      "value": 0.09090909090909091,
      "lo": 0.016232172885884855,
      "hi": 0.37735843645159567,
      "n": 11,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "coverage",
      "stratum": "provenance=stripped",
      "target": "self",
      "value": 0.15,
      "lo": 0.052368745896216595,
      "hi": 0.36041886474075696,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "false_alarm_rate",
      "stratum": "source=commons",
      "target": "self",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "checks",
      "metric": "coverage",
      "stratum": "source=commons",
      "target": "self",
      "value": 0.20512820512820512,
      "lo": 0.06814495768843681,
      "hi": 0.47662923626403914,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "coverage",
      "stratum": "source=edited:sharp-patch",
      "target": "self",
      "value": 0.2333333333333333,
      "lo": 0.07255530696407597,
      "hi": 0.5421288529627982,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "false_alarm_rate",
      "stratum": "source=sroie",
      "target": "self",
      "value": 0.00505050505050505,
      "lo": 0.0003756230402271882,
      "hi": 0.06417236269034904,
      "n": 66,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "checks",
      "metric": "coverage",
      "stratum": "source=sroie",
      "target": "self",
      "value": 0.23039215686274525,
      "lo": 0.14637614552531364,
      "hi": 0.34324075161786194,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "all",
      "target": "self",
      "value": 0.8699074074074075,
      "lo": 0.7888850308641975,
      "hi": 0.941832561728395,
      "n": 121,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "all",
      "target": "self",
      "value": 0.5,
      "lo": 0.35,
      "hi": 0.775,
      "n": 40,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "all",
      "target": "self",
      "value": 0.775,
      "lo": 0.525,
      "hi": 0.9,
      "n": 40,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "all",
      "target": "self",
      "value": 0.8,
      "lo": 0.675,
      "hi": 0.9,
      "n": 40,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "all",
      "target": "self",
      "value": 0.1916666666666666,
      "lo": 0.09907913275458552,
      "hi": 0.33828752131707446,
      "n": 40,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_5",
      "stratum": "all",
      "target": "self",
      "value": 0.008230452674897118,
      "lo": 0.001079429439750073,
      "hi": 0.05991424306373648,
      "n": 81,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "all",
      "target": "self",
      "value": 0.17499999999999993,
      "lo": 0.08745413746035914,
      "hi": 0.3194999033178771,
      "n": 40,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_7",
      "stratum": "all",
      "target": "self",
      "value": 0.004115226337448559,
      "lo": 0.0003060448651870523,
      "hi": 0.052829834102763824,
      "n": 81,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "base=sroie",
      "target": "self",
      "value": 0.15,
      "lo": 0.052368745896216595,
      "hi": 0.36041886474075696,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "base=sroie",
      "target": "self",
      "value": 0.11666666666666665,
      "lo": 0.035533746172033476,
      "hi": 0.3213288750017947,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "capture=photo",
      "target": "self",
      "value": 0.8796296296296297,
      "lo": 0.7222222222222222,
      "hi": 1,
      "n": 21,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "capture=photo",
      "target": "self",
      "value": 0.6666666666666666,
      "lo": 0.4166666666666667,
      "hi": 1,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "capture=photo",
      "target": "self",
      "value": 0.6666666666666666,
      "lo": 0.4166666666666667,
      "hi": 1,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "capture=photo",
      "target": "self",
      "value": 0.6666666666666666,
      "lo": 0.4166666666666667,
      "hi": 1,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "capture=photo",
      "target": "self",
      "value": 0.2222222222222222,
      "lo": 0.07405942055685402,
      "hi": 0.5051039164738353,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "capture=photo",
      "target": "self",
      "value": 0.2222222222222222,
      "lo": 0.07405942055685402,
      "hi": 0.5051039164738353,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "capture=scan",
      "target": "self",
      "value": 0.8626760563380281,
      "lo": 0.7608023138832998,
      "hi": 0.9507168008048289,
      "n": 99,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "capture=scan",
      "target": "self",
      "value": 0.4642857142857143,
      "lo": 0.2857142857142857,
      "hi": 0.7857142857142857,
      "n": 28,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "capture=scan",
      "target": "self",
      "value": 0.75,
      "lo": 0.42857142857142855,
      "hi": 0.8928571428571429,
      "n": 28,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "capture=scan",
      "target": "self",
      "value": 0.7857142857142857,
      "lo": 0.6071428571428571,
      "hi": 0.9285714285714286,
      "n": 28,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "capture=scan",
      "target": "self",
      "value": 0.17857142857142855,
      "lo": 0.07878501945604274,
      "hi": 0.3559142478934294,
      "n": 28,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_5",
      "stratum": "capture=scan",
      "target": "self",
      "value": 0.009389671361502348,
      "lo": 0.0012316058715024641,
      "hi": 0.06791178384281717,
      "n": 71,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "capture=scan",
      "target": "self",
      "value": 0.15476190476190474,
      "lo": 0.0640637580066869,
      "hi": 0.3287613809983017,
      "n": 28,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_7",
      "stratum": "capture=scan",
      "target": "self",
      "value": 0.004694835680751174,
      "lo": 0.0003491627506430964,
      "hi": 0.059886508204961886,
      "n": 71,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "country=MY",
      "target": "self",
      "value": 0.845,
      "lo": 0.7299926470588235,
      "hi": 0.9382426470588234,
      "n": 93,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "country=MY",
      "target": "self",
      "value": 0.44,
      "lo": 0.24,
      "hi": 0.8,
      "n": 25,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "country=MY",
      "target": "self",
      "value": 0.72,
      "lo": 0.4,
      "hi": 0.88,
      "n": 25,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "country=MY",
      "target": "self",
      "value": 0.76,
      "lo": 0.56,
      "hi": 0.92,
      "n": 25,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "country=MY",
      "target": "self",
      "value": 0.1733333333333333,
      "lo": 0.07200481142187823,
      "hi": 0.36168079252496677,
      "n": 25,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_5",
      "stratum": "country=MY",
      "target": "self",
      "value": 0.00980392156862745,
      "lo": 0.0012859952143584458,
      "hi": 0.07074472686414209,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "country=MY",
      "target": "self",
      "value": 0.14666666666666667,
      "lo": 0.05630699669747266,
      "hi": 0.33114886063278826,
      "n": 25,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_7",
      "stratum": "country=MY",
      "target": "self",
      "value": 0.004901960784313725,
      "lo": 0.0003645718312947112,
      "hi": 0.06238645746799082,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "currency=MYR",
      "target": "self",
      "value": 0.845,
      "lo": 0.7349632352941177,
      "hi": 0.9435367647058823,
      "n": 93,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "currency=MYR",
      "target": "self",
      "value": 0.44,
      "lo": 0.28,
      "hi": 0.76,
      "n": 25,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "currency=MYR",
      "target": "self",
      "value": 0.72,
      "lo": 0.36,
      "hi": 0.88,
      "n": 25,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "currency=MYR",
      "target": "self",
      "value": 0.76,
      "lo": 0.56,
      "hi": 0.92,
      "n": 25,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "currency=MYR",
      "target": "self",
      "value": 0.1733333333333333,
      "lo": 0.07200481142187823,
      "hi": 0.36168079252496677,
      "n": 25,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_5",
      "stratum": "currency=MYR",
      "target": "self",
      "value": 0.00980392156862745,
      "lo": 0.0012859952143584458,
      "hi": 0.07074472686414209,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "currency=MYR",
      "target": "self",
      "value": 0.14666666666666667,
      "lo": 0.05630699669747266,
      "hi": 0.33114886063278826,
      "n": 25,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_7",
      "stratum": "currency=MYR",
      "target": "self",
      "value": 0.004901960784313725,
      "lo": 0.0003645718312947112,
      "hi": 0.06238645746799082,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "dateEra=old",
      "target": "self",
      "value": 0.8693627450980392,
      "lo": 0.7720526960784314,
      "hi": 0.9487806372549019,
      "n": 98,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "dateEra=old",
      "target": "self",
      "value": 0.4666666666666667,
      "lo": 0.3,
      "hi": 0.8333333333333334,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "dateEra=old",
      "target": "self",
      "value": 0.7666666666666667,
      "lo": 0.43333333333333335,
      "hi": 0.9,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "dateEra=old",
      "target": "self",
      "value": 0.8,
      "lo": 0.6333333333333333,
      "hi": 0.9333333333333333,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "dateEra=old",
      "target": "self",
      "value": 0.18888888888888883,
      "lo": 0.08768112180333776,
      "hi": 0.3607272117271315,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_5",
      "stratum": "dateEra=old",
      "target": "self",
      "value": 0.00980392156862745,
      "lo": 0.0012859952143584458,
      "hi": 0.07074472686414209,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "dateEra=old",
      "target": "self",
      "value": 0.16666666666666663,
      "lo": 0.07336542371848548,
      "hi": 0.33564350506416024,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_7",
      "stratum": "dateEra=old",
      "target": "self",
      "value": 0.004901960784313725,
      "lo": 0.0003645718312947112,
      "hi": 0.06238645746799082,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "dateEra=recent",
      "target": "self",
      "value": 0.8807692307692307,
      "lo": 0.7076923076923077,
      "hi": 1,
      "n": 23,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "dateEra=recent",
      "target": "self",
      "value": 0.7,
      "lo": 0.4,
      "hi": 1,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "dateEra=recent",
      "target": "self",
      "value": 0.7,
      "lo": 0.4,
      "hi": 1,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "dateEra=recent",
      "target": "self",
      "value": 0.8,
      "lo": 0.5,
      "hi": 1,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "dateEra=recent",
      "target": "self",
      "value": 0.19999999999999998,
      "lo": 0.056682151454375246,
      "hi": 0.5098375284633582,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_5",
      "stratum": "dateEra=recent",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.22809537235419838,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "dateEra=recent",
      "target": "self",
      "value": 0.19999999999999998,
      "lo": 0.056682151454375246,
      "hi": 0.5098375284633582,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_7",
      "stratum": "dateEra=recent",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.22809537235419838,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "docType=receipt",
      "target": "self",
      "value": 0.8699074074074075,
      "lo": 0.7828587962962963,
      "hi": 0.9422916666666666,
      "n": 121,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "docType=receipt",
      "target": "self",
      "value": 0.5,
      "lo": 0.35,
      "hi": 0.8,
      "n": 40,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "docType=receipt",
      "target": "self",
      "value": 0.775,
      "lo": 0.55,
      "hi": 0.9,
      "n": 40,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "docType=receipt",
      "target": "self",
      "value": 0.8,
      "lo": 0.65,
      "hi": 0.9,
      "n": 40,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "docType=receipt",
      "target": "self",
      "value": 0.1916666666666666,
      "lo": 0.09907913275458552,
      "hi": 0.33828752131707446,
      "n": 40,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_5",
      "stratum": "docType=receipt",
      "target": "self",
      "value": 0.008230452674897118,
      "lo": 0.001079429439750073,
      "hi": 0.05991424306373648,
      "n": 81,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "docType=receipt",
      "target": "self",
      "value": 0.17499999999999993,
      "lo": 0.08745413746035914,
      "hi": 0.3194999033178771,
      "n": 40,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_7",
      "stratum": "docType=receipt",
      "target": "self",
      "value": 0.004115226337448559,
      "lo": 0.0003060448651870523,
      "hi": 0.052829834102763824,
      "n": 81,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "lang=en",
      "target": "self",
      "value": 0.8468599033816425,
      "lo": 0.7434601449275363,
      "hi": 0.9369746376811594,
      "n": 99,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "lang=en",
      "target": "self",
      "value": 0.5,
      "lo": 0.3333333333333333,
      "hi": 0.8,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "lang=en",
      "target": "self",
      "value": 0.7333333333333333,
      "lo": 0.43333333333333335,
      "hi": 0.8666666666666667,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "lang=en",
      "target": "self",
      "value": 0.7666666666666667,
      "lo": 0.6,
      "hi": 0.9,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "lang=en",
      "target": "self",
      "value": 0.18888888888888883,
      "lo": 0.08768112180333776,
      "hi": 0.3607272117271315,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_5",
      "stratum": "lang=en",
      "target": "self",
      "value": 0.00966183574879227,
      "lo": 0.0012673394193254278,
      "hi": 0.06977451281438231,
      "n": 69,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "lang=en",
      "target": "self",
      "value": 0.16666666666666663,
      "lo": 0.07336542371848548,
      "hi": 0.33564350506416024,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_7",
      "stratum": "lang=en",
      "target": "self",
      "value": 0.004830917874396135,
      "lo": 0.00035928654494127993,
      "hi": 0.061530268666438445,
      "n": 69,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_5",
      "stratum": "source=commons",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.22809537235419838,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_7",
      "stratum": "source=commons",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.22809537235419838,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "source=edited:sharp-patch",
      "target": "self",
      "value": 0.13333333333333333,
      "lo": 0.029119713490926458,
      "hi": 0.4410710064085256,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "source=edited:sharp-patch",
      "target": "self",
      "value": 0.06666666666666667,
      "lo": 0.008795570298114924,
      "hi": 0.3650661895830557,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_5",
      "stratum": "source=sroie",
      "target": "self",
      "value": 0.00980392156862745,
      "lo": 0.0012859952143584458,
      "hi": 0.07074472686414209,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_7",
      "stratum": "source=sroie",
      "target": "self",
      "value": 0.004901960784313725,
      "lo": 0.0003645718312947112,
      "hi": 0.06238645746799082,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "provenance=intact",
      "target": "self",
      "value": 0.28333333333333327,
      "lo": 0.1340262550429334,
      "hi": 0.5024613134466216,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "provenance=intact",
      "target": "self",
      "value": 0.2666666666666666,
      "lo": 0.12281731303990046,
      "hi": 0.4857077607180819,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "provenance=stripped",
      "target": "self",
      "value": 0.09999999999999999,
      "lo": 0.027866481213768224,
      "hi": 0.3010336452284873,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "provenance=stripped",
      "target": "self",
      "value": 0.08333333333333333,
      "lo": 0.02079482151120035,
      "hi": 0.2801428101994825,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_5",
      "stratum": "provenance=none",
      "target": "self",
      "value": 0.008230452674897118,
      "lo": 0.001079429439750073,
      "hi": 0.05991424306373648,
      "n": 81,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_7",
      "stratum": "provenance=none",
      "target": "self",
      "value": 0.004115226337448559,
      "lo": 0.0003060448651870523,
      "hi": 0.052829834102763824,
      "n": 81,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "class=ai_full",
      "target": "self",
      "value": 0.8959876543209877,
      "lo": 0.7771296296296296,
      "hi": 0.9935185185185185,
      "n": 101,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "class=ai_full",
      "target": "self",
      "value": 0.65,
      "lo": 0.45,
      "hi": 0.9,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "class=ai_full",
      "target": "self",
      "value": 0.85,
      "lo": 0.6,
      "hi": 1,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "class=ai_full",
      "target": "self",
      "value": 0.85,
      "lo": 0.7,
      "hi": 1,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "class=ai_full",
      "target": "self",
      "value": 0.2333333333333333,
      "lo": 0.1011725498384442,
      "hi": 0.4514275344563928,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "class=ai_full",
      "target": "self",
      "value": 0.2333333333333333,
      "lo": 0.1011725498384442,
      "hi": 0.4514275344563928,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "class=ai_inpaint",
      "target": "self",
      "value": 0.7722222222222223,
      "lo": 0.5777469135802469,
      "hi": 0.9333487654320987,
      "n": 91,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "class=ai_inpaint",
      "target": "self",
      "value": 0.5,
      "lo": 0.2,
      "hi": 0.8,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "class=ai_inpaint",
      "target": "self",
      "value": 0.5,
      "lo": 0.2,
      "hi": 0.8,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "class=ai_inpaint",
      "target": "self",
      "value": 0.6,
      "lo": 0.3,
      "hi": 0.9,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "class=ai_inpaint",
      "target": "self",
      "value": 0.16666666666666666,
      "lo": 0.042141900169843094,
      "hi": 0.4762132997387497,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "class=ai_inpaint",
      "target": "self",
      "value": 0.16666666666666666,
      "lo": 0.042141900169843094,
      "hi": 0.4762132997387497,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "class=manual_edit",
      "target": "self",
      "value": 0.9154320987654321,
      "lo": 0.7777777777777778,
      "hi": 0.9975308641975309,
      "n": 91,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "class=manual_edit",
      "target": "self",
      "value": 0.2,
      "lo": 0,
      "hi": 0.9,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "class=manual_edit",
      "target": "self",
      "value": 0.9,
      "lo": 0.3,
      "hi": 1,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "class=manual_edit",
      "target": "self",
      "value": 0.9,
      "lo": 0.7,
      "hi": 1,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "class=manual_edit",
      "target": "self",
      "value": 0.13333333333333333,
      "lo": 0.029119713490926458,
      "hi": 0.4410710064085256,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "class=manual_edit",
      "target": "self",
      "value": 0.06666666666666667,
      "lo": 0.008795570298114924,
      "hi": 0.3650661895830557,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "arithmetic=consistent",
      "target": "self",
      "value": 0.8148148148148148,
      "lo": 0.6815305836139169,
      "hi": 0.9343434343434344,
      "n": 103,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "arithmetic=consistent",
      "target": "self",
      "value": 0.5,
      "lo": 0.3181818181818182,
      "hi": 0.8181818181818182,
      "n": 22,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "arithmetic=consistent",
      "target": "self",
      "value": 0.7272727272727273,
      "lo": 0.5,
      "hi": 0.9090909090909091,
      "n": 22,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "arithmetic=consistent",
      "target": "self",
      "value": 0.7272727272727273,
      "lo": 0.5,
      "hi": 0.9090909090909091,
      "n": 22,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "arithmetic=consistent",
      "target": "self",
      "value": 0.18181818181818182,
      "lo": 0.07306884583422052,
      "hi": 0.38516607437126543,
      "n": 22,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "arithmetic=consistent",
      "target": "self",
      "value": 0.18181818181818182,
      "lo": 0.07306884583422052,
      "hi": 0.38516607437126543,
      "n": 22,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "arithmetic=inconsistent",
      "target": "self",
      "value": 0.9372427983539094,
      "lo": 0.8566529492455418,
      "hi": 0.9924639917695472,
      "n": 99,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "arithmetic=inconsistent",
      "target": "self",
      "value": 0.5,
      "lo": 0.2777777777777778,
      "hi": 0.8333333333333334,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "arithmetic=inconsistent",
      "target": "self",
      "value": 0.8333333333333334,
      "lo": 0.5,
      "hi": 1,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "arithmetic=inconsistent",
      "target": "self",
      "value": 0.8888888888888888,
      "lo": 0.6666666666666666,
      "hi": 1,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "arithmetic=inconsistent",
      "target": "self",
      "value": 0.20370370370370372,
      "lo": 0.07905406745062382,
      "hi": 0.43257806508072894,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "arithmetic=inconsistent",
      "target": "self",
      "value": 0.16666666666666666,
      "lo": 0.05836576805861446,
      "hi": 0.39222038103915724,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "duplicates",
      "metric": "recall",
      "stratum": "all",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "duplicates",
      "metric": "recall_exact",
      "stratum": "all",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "duplicates",
      "metric": "false_match_rate",
      "stratum": "all",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.14865487460862087,
      "n": 22,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "duplicates",
      "metric": "false_match_rate_genuine",
      "stratum": "all",
      "target": "self",
      "value": 0.005649717514124294,
      "lo": 0.0004202046315038635,
      "hi": 0.07131787357294792,
      "n": 59,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "duplicates",
      "metric": "recall",
      "stratum": "arithmetic=consistent",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "duplicates",
      "metric": "recall_exact",
      "stratum": "arithmetic=consistent",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "duplicates",
      "metric": "false_match_rate",
      "stratum": "arithmetic=consistent",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.14865487460862087,
      "n": 22,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "duplicates",
      "metric": "false_match_rate_genuine",
      "stratum": "arithmetic=consistent",
      "target": "self",
      "value": 0.005952380952380952,
      "lo": 0.0004427241870828963,
      "hi": 0.0748917609040762,
      "n": 56,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "duplicates",
      "metric": "recall",
      "stratum": "capture=scan",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.154639018924847,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "duplicates",
      "metric": "recall_exact",
      "stratum": "capture=scan",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.154639018924847,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "duplicates",
      "metric": "false_match_rate",
      "stratum": "capture=scan",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.14865487460862087,
      "n": 22,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "duplicates",
      "metric": "false_match_rate_genuine",
      "stratum": "capture=scan",
      "target": "self",
      "value": 0.006802721088435374,
      "lo": 0.0005059980953141668,
      "hi": 0.08480817732626389,
      "n": 49,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "duplicates",
      "metric": "recall",
      "stratum": "class=duplicate",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "duplicates",
      "metric": "recall_exact",
      "stratum": "class=duplicate",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "duplicates",
      "metric": "false_match_rate",
      "stratum": "class=genuine",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.2588329669680317,
      "n": 11,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "duplicates",
      "metric": "false_match_rate_genuine",
      "stratum": "class=genuine",
      "target": "self",
      "value": 0.005649717514124294,
      "lo": 0.0004202046315038635,
      "hi": 0.07131787357294792,
      "n": 59,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "duplicates",
      "metric": "false_match_rate",
      "stratum": "class=hard_negative",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.2588329669680317,
      "n": 11,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "duplicates",
      "metric": "recall",
      "stratum": "country=MY",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "duplicates",
      "metric": "recall_exact",
      "stratum": "country=MY",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "duplicates",
      "metric": "false_match_rate",
      "stratum": "country=MY",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.14865487460862087,
      "n": 22,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "duplicates",
      "metric": "false_match_rate_genuine",
      "stratum": "country=MY",
      "target": "self",
      "value": 0.007246376811594202,
      "lo": 0.000539013310440345,
      "hi": 0.08991029523766786,
      "n": 46,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "duplicates",
      "metric": "recall",
      "stratum": "currency=MYR",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "duplicates",
      "metric": "recall_exact",
      "stratum": "currency=MYR",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "duplicates",
      "metric": "false_match_rate",
      "stratum": "currency=MYR",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.14865487460862087,
      "n": 22,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "duplicates",
      "metric": "false_match_rate_genuine",
      "stratum": "currency=MYR",
      "target": "self",
      "value": 0.007246376811594202,
      "lo": 0.000539013310440345,
      "hi": 0.08991029523766786,
      "n": 46,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "duplicates",
      "metric": "recall",
      "stratum": "dateEra=old",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "duplicates",
      "metric": "recall_exact",
      "stratum": "dateEra=old",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "duplicates",
      "metric": "false_match_rate",
      "stratum": "dateEra=old",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.14865487460862087,
      "n": 22,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "duplicates",
      "metric": "false_match_rate_genuine",
      "stratum": "dateEra=old",
      "target": "self",
      "value": 0.007246376811594202,
      "lo": 0.000539013310440345,
      "hi": 0.08991029523766786,
      "n": 46,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "duplicates",
      "metric": "false_match_rate_genuine",
      "stratum": "dateEra=recent",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.22809537235419838,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "duplicates",
      "metric": "recall",
      "stratum": "docType=receipt",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "duplicates",
      "metric": "recall_exact",
      "stratum": "docType=receipt",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "duplicates",
      "metric": "false_match_rate",
      "stratum": "docType=receipt",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.14865487460862087,
      "n": 22,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "duplicates",
      "metric": "false_match_rate_genuine",
      "stratum": "docType=receipt",
      "target": "self",
      "value": 0.005649717514124294,
      "lo": 0.0004202046315038635,
      "hi": 0.07131787357294792,
      "n": 59,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "duplicates",
      "metric": "recall",
      "stratum": "lang=en",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "duplicates",
      "metric": "recall_exact",
      "stratum": "lang=en",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "duplicates",
      "metric": "false_match_rate",
      "stratum": "lang=en",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.14865487460862087,
      "n": 22,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "duplicates",
      "metric": "false_match_rate_genuine",
      "stratum": "lang=en",
      "target": "self",
      "value": 0.0070921985815602835,
      "lo": 0.0005275397246634964,
      "hi": 0.0881427225362974,
      "n": 47,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "duplicates",
      "metric": "recall",
      "stratum": "provenance=none",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "duplicates",
      "metric": "recall_exact",
      "stratum": "provenance=none",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "duplicates",
      "metric": "false_match_rate",
      "stratum": "provenance=none",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.14865487460862087,
      "n": 22,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "duplicates",
      "metric": "false_match_rate_genuine",
      "stratum": "provenance=none",
      "target": "self",
      "value": 0.005649717514124294,
      "lo": 0.0004202046315038635,
      "hi": 0.07131787357294792,
      "n": 59,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "duplicates",
      "metric": "false_match_rate_genuine",
      "stratum": "source=commons",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.22809537235419838,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "duplicates",
      "metric": "recall",
      "stratum": "source=sroie",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "duplicates",
      "metric": "recall_exact",
      "stratum": "source=sroie",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "duplicates",
      "metric": "false_match_rate",
      "stratum": "source=sroie",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.14865487460862087,
      "n": 22,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "duplicates",
      "metric": "false_match_rate_genuine",
      "stratum": "source=sroie",
      "target": "self",
      "value": 0.007246376811594202,
      "lo": 0.000539013310440345,
      "hi": 0.08991029523766786,
      "n": 46,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "decision_consistency",
      "stratum": "all",
      "target": "self",
      "value": 0.7933884297520661,
      "lo": 0.7127678369090101,
      "hi": 0.8559534890887507,
      "n": 121,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "decisions",
      "metric": "total_consistency",
      "stratum": "all",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.03077069794747827,
      "n": 121,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "decisions",
      "metric": "calibration_error",
      "stratum": "all",
      "target": "self",
      "value": 0.0922644628099174,
      "lo": 0.06807417355371903,
      "hi": 0.1499551652892562,
      "n": 121,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_5",
      "stratum": "all",
      "target": "self",
      "value": 946.8518518518518,
      "lo": 775.0000000000001,
      "hi": 1171.2962962962963,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_7",
      "stratum": "all",
      "target": "self",
      "value": 894.2592592592594,
      "lo": 783.3333333333335,
      "hi": 1066.1111111111113,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_5",
      "stratum": "assumption=fakes_0_1pct",
      "target": "self",
      "value": 220.6111111111111,
      "lo": 79.16666666666667,
      "hi": 432.7986111111109,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_7",
      "stratum": "assumption=fakes_0_1pct",
      "target": "self",
      "value": 152.38888888888889,
      "lo": 78.33333333333336,
      "hi": 294.6666666666667,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_5",
      "stratum": "assumption=fakes_5pct",
      "target": "self",
      "value": 4174.588477366256,
      "lo": 3875.000000000001,
      "hi": 4508.1198559670775,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_7",
      "stratum": "assumption=fakes_5pct",
      "target": "self",
      "value": 4191.460905349795,
      "lo": 3916.6666666666674,
      "hi": 4474.375,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_5",
      "stratum": "assumption=fake_32usd",
      "target": "self",
      "value": 397.1851851851852,
      "lo": 250.6666666666667,
      "hi": 604.9629629629629,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_7",
      "stratum": "assumption=fake_32usd",
      "target": "self",
      "value": 333.25925925925924,
      "lo": 253.33333333333334,
      "hi": 487.8444444444441,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_5",
      "stratum": "assumption=fake_182usd",
      "target": "self",
      "value": 1609.685185185185,
      "lo": 1425.666666666667,
      "hi": 1832.6296296296296,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_7",
      "stratum": "assumption=fake_182usd",
      "target": "self",
      "value": 1570.7592592592594,
      "lo": 1425.666666666667,
      "hi": 1763.3703703703704,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_5",
      "stratum": "assumption=review_8usd",
      "target": "self",
      "value": 873.5185185185186,
      "lo": 783.314814814815,
      "hi": 988.8749999999991,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_7",
      "stratum": "assumption=review_8usd",
      "target": "self",
      "value": 857.5925925925927,
      "lo": 783.3333333333335,
      "hi": 947.7777777777778,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_5",
      "stratum": "assumption=review_73usd",
      "target": "self",
      "value": 1403.1481481481483,
      "lo": 783.3333333333335,
      "hi": 2303.7037037037035,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_7",
      "stratum": "assumption=review_73usd",
      "target": "self",
      "value": 1122.4074074074076,
      "lo": 791.6666666666667,
      "hi": 1733.8888888888891,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "approve_all_cost_per_1000",
      "stratum": "all",
      "target": "self",
      "value": 1000,
      "lo": 1000,
      "hi": 1000,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "approve_all_cost_per_1000",
      "stratum": "assumption=fakes_0_1pct",
      "target": "self",
      "value": 100,
      "lo": 100,
      "hi": 100,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "approve_all_cost_per_1000",
      "stratum": "assumption=fakes_5pct",
      "target": "self",
      "value": 5000,
      "lo": 5000,
      "hi": 5000,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "approve_all_cost_per_1000",
      "stratum": "assumption=fake_32usd",
      "target": "self",
      "value": 320,
      "lo": 320,
      "hi": 320,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "approve_all_cost_per_1000",
      "stratum": "assumption=fake_182usd",
      "target": "self",
      "value": 1820,
      "lo": 1820,
      "hi": 1820,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "approve_all_cost_per_1000",
      "stratum": "assumption=review_8usd",
      "target": "self",
      "value": 1000,
      "lo": 1000,
      "hi": 1000,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "approve_all_cost_per_1000",
      "stratum": "assumption=review_73usd",
      "target": "self",
      "value": 1000,
      "lo": 1000,
      "hi": 1000,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "evidence",
      "metric": "edited_field_named",
      "stratum": "all",
      "target": "self",
      "value": 0.11666666666666665,
      "lo": 0.035533746172033476,
      "hi": 0.3213288750017947,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "proof",
      "metric": "fakes_caught_with_proof",
      "stratum": "all",
      "target": "self",
      "value": 0.1666666666666666,
      "lo": 0.08175670188843487,
      "hi": 0.3099910322430895,
      "n": 40,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "proof",
      "metric": "fakes_caught_with_proof",
      "stratum": "class=ai_full",
      "target": "self",
      "value": 0.2,
      "lo": 0.08065766257979806,
      "hi": 0.41601743225189364,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "proof",
      "metric": "fakes_caught_with_proof",
      "stratum": "class=ai_inpaint",
      "target": "self",
      "value": 0.16666666666666666,
      "lo": 0.042141900169843094,
      "hi": 0.4762132997387497,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "proof",
      "metric": "fakes_caught_with_proof",
      "stratum": "class=manual_edit",
      "target": "self",
      "value": 0.1,
      "lo": 0.017876213095072896,
      "hi": 0.4041500267952385,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "proof",
      "metric": "genuine_with_proof_rate",
      "stratum": "all",
      "target": "self",
      "value": 0.008230452674897118,
      "lo": 0.001079429439750073,
      "hi": 0.05991424306373648,
      "n": 81,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "all",
      "target": "self",
      "value": 4791,
      "lo": 4791,
      "hi": 4791,
      "n": 121,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "all",
      "target": "self",
      "value": 10400,
      "lo": 10400,
      "hi": 10400,
      "n": 121,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "all",
      "target": "self",
      "value": 12816.999999999996,
      "lo": 12816.999999999996,
      "hi": 12816.999999999996,
      "n": 121,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "all",
      "target": "self",
      "value": 0.01633333333333333,
      "lo": 0.01633333333333333,
      "hi": 0.01633333333333333,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "all",
      "target": "self",
      "value": 0.0010668900826446284,
      "lo": 0.0010668900826446284,
      "hi": 0.0010668900826446284,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "all",
      "target": "self",
      "value": 0.7328918322295805,
      "lo": 0.6903140437600526,
      "hi": 0.77155296923978,
      "n": 453,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "all",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.008408735123582253,
      "n": 453,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "all",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.008408735123582253,
      "n": 453,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "arithmetic=consistent",
      "target": "self",
      "value": 4754.5,
      "lo": 4754.5,
      "hi": 4754.5,
      "n": 100,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "arithmetic=consistent",
      "target": "self",
      "value": 9869.999999999998,
      "lo": 9869.999999999998,
      "hi": 9869.999999999998,
      "n": 100,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "arithmetic=consistent",
      "target": "self",
      "value": 11576.570000000018,
      "lo": 11576.570000000018,
      "hi": 11576.570000000018,
      "n": 100,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "arithmetic=consistent",
      "target": "self",
      "value": 0.01633333333333333,
      "lo": 0.01633333333333333,
      "hi": 0.01633333333333333,
      "n": 100,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "arithmetic=consistent",
      "target": "self",
      "value": 0.001052327,
      "lo": 0.001052327,
      "hi": 0.001052327,
      "n": 100,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "arithmetic=consistent",
      "target": "self",
      "value": 0.7435897435897436,
      "lo": 0.6980241406102549,
      "hi": 0.7844034853916643,
      "n": 390,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "arithmetic=consistent",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.009753820311850517,
      "n": 390,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "arithmetic=consistent",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.009753820311850517,
      "n": 390,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "arithmetic=inconsistent",
      "target": "self",
      "value": 6160,
      "lo": 6160,
      "hi": 6160,
      "n": 18,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "arithmetic=inconsistent",
      "target": "self",
      "value": 10623.949999999997,
      "lo": 10623.949999999997,
      "hi": 10623.949999999997,
      "n": 18,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "arithmetic=inconsistent",
      "target": "self",
      "value": 11639.189999999997,
      "lo": 11639.189999999997,
      "hi": 11639.189999999997,
      "n": 18,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "arithmetic=inconsistent",
      "target": "self",
      "value": 0.01633333333333333,
      "lo": 0.01633333333333333,
      "hi": 0.01633333333333333,
      "n": 18,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "arithmetic=inconsistent",
      "target": "self",
      "value": 0.0011256166666666666,
      "lo": 0.0011256166666666666,
      "hi": 0.0011256166666666666,
      "n": 18,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "arithmetic=inconsistent",
      "target": "self",
      "value": 0.6666666666666666,
      "lo": 0.5336096360112935,
      "hi": 0.7775858346250742,
      "n": 54,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "arithmetic=inconsistent",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.0664135880908964,
      "n": 54,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "arithmetic=inconsistent",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.0664135880908964,
      "n": 54,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "arithmeticError=totals",
      "target": "self",
      "value": 0.6666666666666666,
      "lo": 0.4537345197425524,
      "hi": 0.8280524739491653,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "arithmeticError=totals",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.154639018924847,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "arithmeticError=totals",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.154639018924847,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "base=sroie",
      "target": "self",
      "value": 5544,
      "lo": 5544,
      "hi": 5544,
      "n": 20,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "base=sroie",
      "target": "self",
      "value": 10474.650000000001,
      "lo": 10474.650000000001,
      "hi": 10474.650000000001,
      "n": 20,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "base=sroie",
      "target": "self",
      "value": 11609.329999999998,
      "lo": 11609.329999999998,
      "hi": 11609.329999999998,
      "n": 20,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "base=sroie",
      "target": "self",
      "value": 0.01633333333333333,
      "lo": 0.01633333333333333,
      "hi": 0.01633333333333333,
      "n": 20,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "base=sroie",
      "target": "self",
      "value": 0.00110333,
      "lo": 0.00110333,
      "hi": 0.00110333,
      "n": 20,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "base=sroie",
      "target": "self",
      "value": 0.6666666666666666,
      "lo": 0.5405686645211968,
      "hi": 0.7727073847647731,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "base=sroie",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.06017185214208986,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "base=sroie",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.06017185214208986,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "capture=photo",
      "target": "self",
      "value": 7537,
      "lo": 7537,
      "hi": 7537,
      "n": 21,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "capture=photo",
      "target": "self",
      "value": 11540,
      "lo": 11540,
      "hi": 11540,
      "n": 21,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "capture=photo",
      "target": "self",
      "value": 14465.600000000002,
      "lo": 14465.600000000002,
      "hi": 14465.600000000002,
      "n": 21,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "capture=photo",
      "target": "self",
      "value": 0.01633333333333333,
      "lo": 0.01633333333333333,
      "hi": 0.01633333333333333,
      "n": 21,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "capture=photo",
      "target": "self",
      "value": 0.0012058857142857146,
      "lo": 0.0012058857142857146,
      "hi": 0.0012058857142857146,
      "n": 21,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "capture=photo",
      "target": "self",
      "value": 0.72,
      "lo": 0.6095930981193548,
      "hi": 0.8089684111882106,
      "n": 75,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "capture=photo",
      "target": "self",
      "value": 0,
      "lo": 3.469446951953614e-18,
      "hi": 0.04872384248280587,
      "n": 75,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "capture=photo",
      "target": "self",
      "value": 0,
      "lo": 3.469446951953614e-18,
      "hi": 0.04872384248280587,
      "n": 75,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "capture=scan",
      "target": "self",
      "value": 4466,
      "lo": 4466,
      "hi": 4466,
      "n": 99,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "capture=scan",
      "target": "self",
      "value": 10374.8,
      "lo": 10374.8,
      "hi": 10374.8,
      "n": 99,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "capture=scan",
      "target": "self",
      "value": 11916.099999999995,
      "lo": 11916.099999999995,
      "hi": 11916.099999999995,
      "n": 99,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "capture=scan",
      "target": "self",
      "value": 0.01633333333333333,
      "lo": 0.01633333333333333,
      "hi": 0.01633333333333333,
      "n": 99,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "capture=scan",
      "target": "self",
      "value": 0.0010355151515151519,
      "lo": 0.0010355151515151519,
      "hi": 0.0010355151515151519,
      "n": 99,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "capture=scan",
      "target": "self",
      "value": 0.725,
      "lo": 0.6766825620663512,
      "hi": 0.7685663123337365,
      "n": 360,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "capture=scan",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.010558056888693507,
      "n": 360,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "capture=scan",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.010558056888693507,
      "n": 360,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "capture=screenshot",
      "target": "self",
      "value": 1,
      "lo": 0.7961166989641515,
      "hi": 1,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "capture=screenshot",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "capture=screenshot",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "class=ai_full",
      "target": "self",
      "value": 6533,
      "lo": 6533,
      "hi": 6533,
      "n": 20,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "class=ai_full",
      "target": "self",
      "value": 9547.350000000004,
      "lo": 9547.350000000004,
      "hi": 9547.350000000004,
      "n": 20,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "class=ai_full",
      "target": "self",
      "value": 14067.069999999992,
      "lo": 14067.069999999992,
      "hi": 14067.069999999992,
      "n": 20,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "class=ai_full",
      "target": "self",
      "value": 0.01633333333333333,
      "lo": 0.01633333333333333,
      "hi": 0.01633333333333333,
      "n": 20,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "class=ai_full",
      "target": "self",
      "value": 0.0011121599999999998,
      "lo": 0.0011121599999999998,
      "hi": 0.0011121599999999998,
      "n": 20,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "class=ai_full",
      "target": "self",
      "value": 0.6666666666666666,
      "lo": 0.5405686645211968,
      "hi": 0.7727073847647731,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "class=ai_full",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.06017185214208986,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "class=ai_full",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.06017185214208986,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "class=ai_inpaint",
      "target": "self",
      "value": 5153,
      "lo": 5153,
      "hi": 5153,
      "n": 10,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "class=ai_inpaint",
      "target": "self",
      "value": 8857.749999999996,
      "lo": 8857.749999999996,
      "hi": 8857.749999999996,
      "n": 10,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "class=ai_inpaint",
      "target": "self",
      "value": 10069.150000000001,
      "lo": 10069.150000000001,
      "hi": 10069.150000000001,
      "n": 10,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "class=ai_inpaint",
      "target": "self",
      "value": 0.01633333333333333,
      "lo": 0.01633333333333333,
      "hi": 0.01633333333333333,
      "n": 10,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "class=ai_inpaint",
      "target": "self",
      "value": 0.0010525399999999998,
      "lo": 0.0010525399999999998,
      "hi": 0.0010525399999999998,
      "n": 10,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "class=ai_inpaint",
      "target": "self",
      "value": 0.6666666666666666,
      "lo": 0.4878005164454384,
      "hi": 0.8076950191632386,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "class=ai_inpaint",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "class=ai_inpaint",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "class=duplicate",
      "target": "self",
      "value": 1,
      "lo": 0.9590643744356734,
      "hi": 1,
      "n": 90,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "class=duplicate",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.04093562556432678,
      "n": 90,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "class=duplicate",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.04093562556432678,
      "n": 90,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "class=genuine",
      "target": "self",
      "value": 4767,
      "lo": 4767,
      "hi": 4767,
      "n": 70,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "class=genuine",
      "target": "self",
      "value": 10579.849999999999,
      "lo": 10579.849999999999,
      "hi": 10579.849999999999,
      "n": 70,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "class=genuine",
      "target": "self",
      "value": 12007.480000000003,
      "lo": 12007.480000000003,
      "hi": 12007.480000000003,
      "n": 70,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "class=genuine",
      "target": "self",
      "value": 0.01633333333333333,
      "lo": 0.01633333333333333,
      "hi": 0.01633333333333333,
      "n": 70,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "class=genuine",
      "target": "self",
      "value": 0.0010621614285714284,
      "lo": 0.0010621614285714284,
      "hi": 0.0010621614285714284,
      "n": 70,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "class=genuine",
      "target": "self",
      "value": 0.6666666666666666,
      "lo": 0.6004193945242017,
      "hi": 0.7269259218893706,
      "n": 210,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "class=genuine",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.017964050759283236,
      "n": 210,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "class=genuine",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.017964050759283236,
      "n": 210,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "class=hard_negative",
      "target": "self",
      "value": 4355,
      "lo": 4355,
      "hi": 4355,
      "n": 11,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "class=hard_negative",
      "target": "self",
      "value": 7242.5,
      "lo": 7242.5,
      "hi": 7242.5,
      "n": 11,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "class=hard_negative",
      "target": "self",
      "value": 8960.500000000002,
      "lo": 8960.500000000002,
      "hi": 8960.500000000002,
      "n": 11,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "class=hard_negative",
      "target": "self",
      "value": 0.01633333333333333,
      "lo": 0.01633333333333333,
      "hi": 0.01633333333333333,
      "n": 11,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "class=hard_negative",
      "target": "self",
      "value": 0.0009484181818181818,
      "lo": 0.0009484181818181818,
      "hi": 0.0009484181818181818,
      "n": 11,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "class=hard_negative",
      "target": "self",
      "value": 0.6666666666666666,
      "lo": 0.4960789442780561,
      "hi": 0.8024977232077931,
      "n": 33,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "class=hard_negative",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.10426999754245206,
      "n": 33,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "class=hard_negative",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.10426999754245206,
      "n": 33,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "class=manual_edit",
      "target": "self",
      "value": 6219,
      "lo": 6219,
      "hi": 6219,
      "n": 10,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "class=manual_edit",
      "target": "self",
      "value": 11221.149999999998,
      "lo": 11221.149999999998,
      "hi": 11221.149999999998,
      "n": 10,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "class=manual_edit",
      "target": "self",
      "value": 11758.630000000001,
      "lo": 11758.630000000001,
      "hi": 11758.630000000001,
      "n": 10,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "class=manual_edit",
      "target": "self",
      "value": 0.01633333333333333,
      "lo": 0.01633333333333333,
      "hi": 0.01633333333333333,
      "n": 10,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "class=manual_edit",
      "target": "self",
      "value": 0.0011541200000000002,
      "lo": 0.0011541200000000002,
      "hi": 0.0011541200000000002,
      "n": 10,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "class=manual_edit",
      "target": "self",
      "value": 0.6666666666666666,
      "lo": 0.4878005164454384,
      "hi": 0.8076950191632386,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "class=manual_edit",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "class=manual_edit",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "country=JP",
      "target": "self",
      "value": 0.6666666666666666,
      "lo": 0.39062208887279953,
      "hi": 0.8618799089087867,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "country=JP",
      "target": "self",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "country=JP",
      "target": "self",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "country=MY",
      "target": "self",
      "value": 4479,
      "lo": 4479,
      "hi": 4479,
      "n": 93,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "country=MY",
      "target": "self",
      "value": 10383.199999999999,
      "lo": 10383.199999999999,
      "hi": 10383.199999999999,
      "n": 93,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "country=MY",
      "target": "self",
      "value": 11985.399999999998,
      "lo": 11985.399999999998,
      "hi": 11985.399999999998,
      "n": 93,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "country=MY",
      "target": "self",
      "value": 0.01633333333333333,
      "lo": 0.01633333333333333,
      "hi": 0.01633333333333333,
      "n": 93,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "country=MY",
      "target": "self",
      "value": 0.0010424182795698929,
      "lo": 0.0010424182795698929,
      "hi": 0.0010424182795698929,
      "n": 93,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "country=MY",
      "target": "self",
      "value": 0.7479674796747967,
      "lo": 0.7012673526138609,
      "hi": 0.7895578913888663,
      "n": 369,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "country=MY",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.010303196519090837,
      "n": 369,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "country=MY",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.010303196519090837,
      "n": 369,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "currency=JPY",
      "target": "self",
      "value": 0.6666666666666666,
      "lo": 0.39062208887279953,
      "hi": 0.8618799089087867,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "currency=JPY",
      "target": "self",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "currency=JPY",
      "target": "self",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "currency=MYR",
      "target": "self",
      "value": 4479,
      "lo": 4479,
      "hi": 4479,
      "n": 93,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "currency=MYR",
      "target": "self",
      "value": 10383.199999999999,
      "lo": 10383.199999999999,
      "hi": 10383.199999999999,
      "n": 93,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "currency=MYR",
      "target": "self",
      "value": 11985.399999999998,
      "lo": 11985.399999999998,
      "hi": 11985.399999999998,
      "n": 93,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "currency=MYR",
      "target": "self",
      "value": 0.01633333333333333,
      "lo": 0.01633333333333333,
      "hi": 0.01633333333333333,
      "n": 93,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "currency=MYR",
      "target": "self",
      "value": 0.0010424182795698929,
      "lo": 0.0010424182795698929,
      "hi": 0.0010424182795698929,
      "n": 93,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "currency=MYR",
      "target": "self",
      "value": 0.7479674796747967,
      "lo": 0.7012673526138609,
      "hi": 0.7895578913888663,
      "n": 369,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "currency=MYR",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.010303196519090837,
      "n": 369,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "currency=MYR",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.010303196519090837,
      "n": 369,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "dateEra=old",
      "target": "self",
      "value": 4472.5,
      "lo": 4472.5,
      "hi": 4472.5,
      "n": 98,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "dateEra=old",
      "target": "self",
      "value": 10376.199999999999,
      "lo": 10376.199999999999,
      "hi": 10376.199999999999,
      "n": 98,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "dateEra=old",
      "target": "self",
      "value": 11927.650000000001,
      "lo": 11927.650000000001,
      "hi": 11927.650000000001,
      "n": 98,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "dateEra=old",
      "target": "self",
      "value": 0.01633333333333333,
      "lo": 0.01633333333333333,
      "hi": 0.01633333333333333,
      "n": 98,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "dateEra=old",
      "target": "self",
      "value": 0.0010386642857142863,
      "lo": 0.0010386642857142863,
      "hi": 0.0010386642857142863,
      "n": 98,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "dateEra=old",
      "target": "self",
      "value": 0.7447916666666666,
      "lo": 0.6989097695782239,
      "hi": 0.785824380740955,
      "n": 384,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "dateEra=old",
      "target": "self",
      "value": 0,
      "lo": 8.673617379884035e-19,
      "hi": 0.009904714241677031,
      "n": 384,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "dateEra=old",
      "target": "self",
      "value": 0,
      "lo": 8.673617379884035e-19,
      "hi": 0.009904714241677031,
      "n": 384,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "dateEra=recent",
      "target": "self",
      "value": 7537,
      "lo": 7537,
      "hi": 7537,
      "n": 23,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "dateEra=recent",
      "target": "self",
      "value": 11370.399999999998,
      "lo": 11370.399999999998,
      "hi": 11370.399999999998,
      "n": 23,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "dateEra=recent",
      "target": "self",
      "value": 14392.460000000005,
      "lo": 14392.460000000005,
      "hi": 14392.460000000005,
      "n": 23,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "dateEra=recent",
      "target": "self",
      "value": 0.01633333333333333,
      "lo": 0.01633333333333333,
      "hi": 0.01633333333333333,
      "n": 23,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "dateEra=recent",
      "target": "self",
      "value": 0.0011871565217391306,
      "lo": 0.0011871565217391306,
      "hi": 0.0011871565217391306,
      "n": 23,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "dateEra=recent",
      "target": "self",
      "value": 0.6666666666666666,
      "lo": 0.5492647542871832,
      "hi": 0.7664894929831328,
      "n": 69,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "dateEra=recent",
      "target": "self",
      "value": 0,
      "lo": 3.469446951953614e-18,
      "hi": 0.05273725818905173,
      "n": 69,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "dateEra=recent",
      "target": "self",
      "value": 0,
      "lo": 3.469446951953614e-18,
      "hi": 0.05273725818905173,
      "n": 69,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "docType=receipt",
      "target": "self",
      "value": 4791,
      "lo": 4791,
      "hi": 4791,
      "n": 121,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "docType=receipt",
      "target": "self",
      "value": 10400,
      "lo": 10400,
      "hi": 10400,
      "n": 121,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "docType=receipt",
      "target": "self",
      "value": 12816.999999999996,
      "lo": 12816.999999999996,
      "hi": 12816.999999999996,
      "n": 121,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "docType=receipt",
      "target": "self",
      "value": 0.01633333333333333,
      "lo": 0.01633333333333333,
      "hi": 0.01633333333333333,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "docType=receipt",
      "target": "self",
      "value": 0.0010668900826446284,
      "lo": 0.0010668900826446284,
      "hi": 0.0010668900826446284,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "docType=receipt",
      "target": "self",
      "value": 0.7328918322295805,
      "lo": 0.6903140437600526,
      "hi": 0.77155296923978,
      "n": 453,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "docType=receipt",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.008408735123582253,
      "n": 453,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "docType=receipt",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.008408735123582253,
      "n": 453,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "editedField=date",
      "target": "self",
      "value": 0.6666666666666666,
      "lo": 0.39062208887279953,
      "hi": 0.8618799089087867,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "editedField=date",
      "target": "self",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "editedField=date",
      "target": "self",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "editedField=items+total",
      "target": "self",
      "value": 0.6666666666666666,
      "lo": 0.39062208887279953,
      "hi": 0.8618799089087867,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "editedField=items+total",
      "target": "self",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "editedField=items+total",
      "target": "self",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "editedField=total",
      "target": "self",
      "value": 0.6666666666666666,
      "lo": 0.47824760507320074,
      "hi": 0.8135673850128435,
      "n": 27,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "editedField=total",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.12455502974186705,
      "n": 27,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "editedField=total",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.12455502974186705,
      "n": 27,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "lang=en",
      "target": "self",
      "value": 4541,
      "lo": 4541,
      "hi": 4541,
      "n": 99,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "lang=en",
      "target": "self",
      "value": 10437.199999999997,
      "lo": 10437.199999999997,
      "hi": 10437.199999999997,
      "n": 99,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "lang=en",
      "target": "self",
      "value": 13090.979999999992,
      "lo": 13090.979999999992,
      "hi": 13090.979999999992,
      "n": 99,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "lang=en",
      "target": "self",
      "value": 0.01633333333333333,
      "lo": 0.01633333333333333,
      "hi": 0.01633333333333333,
      "n": 99,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "lang=en",
      "target": "self",
      "value": 0.0010506393939393945,
      "lo": 0.0010506393939393945,
      "hi": 0.0010506393939393945,
      "n": 99,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "lang=en",
      "target": "self",
      "value": 0.7441860465116279,
      "lo": 0.6984630598221151,
      "hi": 0.7851089761903369,
      "n": 387,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "lang=en",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.009828688164979105,
      "n": 387,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "lang=en",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.009828688164979105,
      "n": 387,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "lang=ja",
      "target": "self",
      "value": 0.6666666666666666,
      "lo": 0.39062208887279953,
      "hi": 0.8618799089087867,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "lang=ja",
      "target": "self",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "lang=ja",
      "target": "self",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "lang=zh",
      "target": "self",
      "value": 0.6666666666666666,
      "lo": 0.43749467295945105,
      "hi": 0.837212252491663,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "lang=zh",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.17587922364665765,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "lang=zh",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.17587922364665765,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "prompt=explicit_amounts",
      "target": "self",
      "value": 0.6666666666666666,
      "lo": 0.4670631683813174,
      "hi": 0.8202780967270225,
      "n": 24,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "prompt=explicit_amounts",
      "target": "self",
      "value": 0,
      "lo": 1.3877787807814457e-17,
      "hi": 0.13797620467498017,
      "n": 24,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "prompt=explicit_amounts",
      "target": "self",
      "value": 0,
      "lo": 1.3877787807814457e-17,
      "hi": 0.13797620467498017,
      "n": 24,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "prompt=free",
      "target": "self",
      "value": 0.6666666666666666,
      "lo": 0.41713547738519774,
      "hi": 0.8482367556028526,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "prompt=free",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "prompt=free",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "prompt=target_total",
      "target": "self",
      "value": 0.6666666666666666,
      "lo": 0.4537345197425524,
      "hi": 0.8280524739491653,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "prompt=target_total",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.154639018924847,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "prompt=target_total",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.154639018924847,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "provenance=intact",
      "target": "self",
      "value": 5568,
      "lo": 5568,
      "hi": 5568,
      "n": 20,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "provenance=intact",
      "target": "self",
      "value": 10448.050000000001,
      "lo": 10448.050000000001,
      "hi": 10448.050000000001,
      "n": 20,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "provenance=intact",
      "target": "self",
      "value": 11604.009999999998,
      "lo": 11604.009999999998,
      "hi": 11604.009999999998,
      "n": 20,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "provenance=intact",
      "target": "self",
      "value": 0.01633333333333333,
      "lo": 0.01633333333333333,
      "hi": 0.01633333333333333,
      "n": 20,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "provenance=intact",
      "target": "self",
      "value": 0.0011166050000000001,
      "lo": 0.0011166050000000001,
      "hi": 0.0011166050000000001,
      "n": 20,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "provenance=intact",
      "target": "self",
      "value": 0.6666666666666666,
      "lo": 0.5405686645211968,
      "hi": 0.7727073847647731,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "provenance=intact",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.06017185214208986,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "provenance=intact",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.06017185214208986,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "provenance=none",
      "target": "self",
      "value": 4541,
      "lo": 4541,
      "hi": 4541,
      "n": 81,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "provenance=none",
      "target": "self",
      "value": 10345,
      "lo": 10345,
      "hi": 10345,
      "n": 81,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "provenance=none",
      "target": "self",
      "value": 11841.600000000004,
      "lo": 11841.600000000004,
      "hi": 11841.600000000004,
      "n": 81,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "provenance=none",
      "target": "self",
      "value": 0.01633333333333333,
      "lo": 0.01633333333333333,
      "hi": 0.01633333333333333,
      "n": 81,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "provenance=none",
      "target": "self",
      "value": 0.001046714814814815,
      "lo": 0.001046714814814815,
      "hi": 0.001046714814814815,
      "n": 81,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "provenance=none",
      "target": "self",
      "value": 0.7567567567567568,
      "lo": 0.7079173792646796,
      "hi": 0.7997398449935313,
      "n": 333,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "provenance=none",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.011404352760326315,
      "n": 333,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "provenance=none",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.011404352760326315,
      "n": 333,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "provenance=stripped",
      "target": "self",
      "value": 6203,
      "lo": 6203,
      "hi": 6203,
      "n": 20,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "provenance=stripped",
      "target": "self",
      "value": 10639.850000000004,
      "lo": 10639.850000000004,
      "hi": 10639.850000000004,
      "n": 20,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "provenance=stripped",
      "target": "self",
      "value": 14285.569999999994,
      "lo": 14285.569999999994,
      "hi": 14285.569999999994,
      "n": 20,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "provenance=stripped",
      "target": "self",
      "value": 0.01633333333333333,
      "lo": 0.01633333333333333,
      "hi": 0.01633333333333333,
      "n": 20,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "provenance=stripped",
      "target": "self",
      "value": 0.0010988849999999997,
      "lo": 0.0010988849999999997,
      "hi": 0.0010988849999999997,
      "n": 20,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "provenance=stripped",
      "target": "self",
      "value": 0.6666666666666666,
      "lo": 0.5405686645211968,
      "hi": 0.7727073847647731,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "provenance=stripped",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.06017185214208986,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "provenance=stripped",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.06017185214208986,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "source=commons",
      "target": "self",
      "value": 6527,
      "lo": 6527,
      "hi": 6527,
      "n": 13,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "source=commons",
      "target": "self",
      "value": 10522.399999999998,
      "lo": 10522.399999999998,
      "hi": 10522.399999999998,
      "n": 13,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "source=commons",
      "target": "self",
      "value": 11336.479999999998,
      "lo": 11336.479999999998,
      "hi": 11336.479999999998,
      "n": 13,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "source=commons",
      "target": "self",
      "value": 0.01633333333333333,
      "lo": 0.01633333333333333,
      "hi": 0.01633333333333333,
      "n": 13,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "source=commons",
      "target": "self",
      "value": 0.001153523076923077,
      "lo": 0.001153523076923077,
      "hi": 0.001153523076923077,
      "n": 13,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "source=commons",
      "target": "self",
      "value": 0.6666666666666666,
      "lo": 0.5097739439934599,
      "hi": 0.7936704381040163,
      "n": 39,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "source=commons",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.08966685370757144,
      "n": 39,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "source=commons",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.08966685370757144,
      "n": 39,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "source=edited:gemini-nano-banana-2.1",
      "target": "self",
      "value": 0.6666666666666666,
      "lo": 0.39062208887279953,
      "hi": 0.8618799089087867,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "source=edited:gemini-nano-banana-2.1",
      "target": "self",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "source=edited:gemini-nano-banana-2.1",
      "target": "self",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "source=edited:gpt-image-2.5-sunburst",
      "target": "self",
      "value": 0.6666666666666666,
      "lo": 0.39062208887279953,
      "hi": 0.8618799089087867,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "source=edited:gpt-image-2.5-sunburst",
      "target": "self",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "source=edited:gpt-image-2.5-sunburst",
      "target": "self",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "source=edited:sharp-patch",
      "target": "self",
      "value": 6219,
      "lo": 6219,
      "hi": 6219,
      "n": 10,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "source=edited:sharp-patch",
      "target": "self",
      "value": 11221.149999999998,
      "lo": 11221.149999999998,
      "hi": 11221.149999999998,
      "n": 10,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "source=edited:sharp-patch",
      "target": "self",
      "value": 11758.630000000001,
      "lo": 11758.630000000001,
      "hi": 11758.630000000001,
      "n": 10,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "source=edited:sharp-patch",
      "target": "self",
      "value": 0.01633333333333333,
      "lo": 0.01633333333333333,
      "hi": 0.01633333333333333,
      "n": 10,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "source=edited:sharp-patch",
      "target": "self",
      "value": 0.0011541200000000002,
      "lo": 0.0011541200000000002,
      "hi": 0.0011541200000000002,
      "n": 10,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "source=edited:sharp-patch",
      "target": "self",
      "value": 0.6666666666666666,
      "lo": 0.4878005164454384,
      "hi": 0.8076950191632386,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "source=edited:sharp-patch",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "source=edited:sharp-patch",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "source=generated:flux-3-image",
      "target": "self",
      "value": 0.6666666666666666,
      "lo": 0.43749467295945105,
      "hi": 0.837212252491663,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "source=generated:flux-3-image",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.17587922364665765,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "source=generated:flux-3-image",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.17587922364665765,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "source=generated:gemini-nano-banana-2.1",
      "target": "self",
      "value": 0.6666666666666666,
      "lo": 0.4537345197425524,
      "hi": 0.8280524739491653,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "source=generated:gemini-nano-banana-2.1",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.154639018924847,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "source=generated:gemini-nano-banana-2.1",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.154639018924847,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "source=generated:gpt-image-2.5-flare",
      "target": "self",
      "value": 0.6666666666666666,
      "lo": 0.4537345197425524,
      "hi": 0.8280524739491653,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "source=generated:gpt-image-2.5-flare",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.154639018924847,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "source=generated:gpt-image-2.5-flare",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.154639018924847,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "source=sroie",
      "target": "self",
      "value": 4456,
      "lo": 4456,
      "hi": 4456,
      "n": 68,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "source=sroie",
      "target": "self",
      "value": 10170,
      "lo": 10170,
      "hi": 10170,
      "n": 68,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "source=sroie",
      "target": "self",
      "value": 11766.959999999997,
      "lo": 11766.959999999997,
      "hi": 11766.959999999997,
      "n": 68,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "source=sroie",
      "target": "self",
      "value": 0.01633333333333333,
      "lo": 0.01633333333333333,
      "hi": 0.01633333333333333,
      "n": 68,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "source=sroie",
      "target": "self",
      "value": 0.0010262955882352944,
      "lo": 0.0010262955882352944,
      "hi": 0.0010262955882352944,
      "n": 68,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "source=sroie",
      "target": "self",
      "value": 0.7687074829931972,
      "lo": 0.7172296222341344,
      "hi": 0.8132539465262468,
      "n": 294,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "source=sroie",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.012897663192708011,
      "n": 294,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "source=sroie",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.012897663192708011,
      "n": 294,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "transform=crop",
      "target": "self",
      "value": 1,
      "lo": 0.7961166989641515,
      "hi": 1,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "transform=crop",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "transform=crop",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "transform=jpeg",
      "target": "self",
      "value": 1,
      "lo": 0.7961166989641515,
      "hi": 1,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "transform=jpeg",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "transform=jpeg",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "transform=rephoto",
      "target": "self",
      "value": 1,
      "lo": 0.7961166989641515,
      "hi": 1,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "transform=rephoto",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "transform=rephoto",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "transform=resize",
      "target": "self",
      "value": 1,
      "lo": 0.7961166989641515,
      "hi": 1,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "transform=resize",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "transform=resize",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "transform=rotate",
      "target": "self",
      "value": 1,
      "lo": 0.7961166989641515,
      "hi": 1,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "transform=rotate",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "transform=rotate",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "transform=screenshot",
      "target": "self",
      "value": 1,
      "lo": 0.7961166989641515,
      "hi": 1,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "transform=screenshot",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "transform=screenshot",
      "target": "self",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "all",
      "target": "self-gemini",
      "value": 0.3298245614035083,
      "lo": 0.24351670349154178,
      "hi": 0.4293601063096113,
      "n": 95,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "all",
      "target": "self-gemini",
      "value": 0.33035714285714274,
      "lo": 0.2501477498939,
      "hi": 0.42181770714667,
      "n": 112,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "all",
      "target": "self-gemini",
      "value": 0.3333333333333329,
      "lo": 0.2436427243338542,
      "hi": 0.4369662978877758,
      "n": 88,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.subtotal",
      "stratum": "all",
      "target": "self-gemini",
      "value": 0.31249999999999994,
      "lo": 0.14164643854782033,
      "hi": 0.5559564416525933,
      "n": 16,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.tax",
      "stratum": "all",
      "target": "self-gemini",
      "value": 0.3333333333333332,
      "lo": 0.16906782991982014,
      "hi": 0.5513072227644528,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "all",
      "target": "self-gemini",
      "value": 0.3333333333333334,
      "lo": 0.255611110856911,
      "hi": 0.4213124551255819,
      "n": 121,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.receipt_number",
      "stratum": "all",
      "target": "self-gemini",
      "value": 0.33333333333333337,
      "lo": 0.132793283187926,
      "hi": 0.6201510391347513,
      "n": 11,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "all",
      "target": "self-gemini",
      "value": 0.325068870523416,
      "lo": 0.24811064057894833,
      "hi": 0.412792606361354,
      "n": 121,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "line_items_f1",
      "stratum": "all",
      "target": "self-gemini",
      "value": 0.43023255813953487,
      "lo": 0.3630695812807882,
      "hi": 0.4827586206896552,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "all",
      "target": "self-gemini",
      "value": 0.3333333333333334,
      "lo": 0.255611110856911,
      "hi": 0.4213124551255819,
      "n": 121,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "arithmetic=consistent",
      "target": "self-gemini",
      "value": 0.33333333333333304,
      "lo": 0.23708220863575497,
      "hi": 0.4458257388585131,
      "n": 75,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "arithmetic=consistent",
      "target": "self-gemini",
      "value": 0.3297101449275358,
      "lo": 0.2421712593668332,
      "hi": 0.43089993843334146,
      "n": 92,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "arithmetic=consistent",
      "target": "self-gemini",
      "value": 0.3333333333333331,
      "lo": 0.23593569195030484,
      "hi": 0.4473949773682717,
      "n": 73,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.subtotal",
      "stratum": "arithmetic=consistent",
      "target": "self-gemini",
      "value": 0.30303030303030304,
      "lo": 0.11471049994616331,
      "hi": 0.5933146082533644,
      "n": 11,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.tax",
      "stratum": "arithmetic=consistent",
      "target": "self-gemini",
      "value": 0.3333333333333333,
      "lo": 0.1430225574285459,
      "hi": 0.5996759000228535,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "arithmetic=consistent",
      "target": "self-gemini",
      "value": 0.3333333333333329,
      "lo": 0.24862102140164288,
      "hi": 0.43037681133401817,
      "n": 100,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "arithmetic=consistent",
      "target": "self-gemini",
      "value": 0.3266666666666662,
      "lo": 0.24264634657098794,
      "hi": 0.42351139947409944,
      "n": 100,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "line_items_f1",
      "stratum": "arithmetic=consistent",
      "target": "self-gemini",
      "value": 0.4375,
      "lo": 0.3389830508474576,
      "hi": 0.5,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "arithmetic=consistent",
      "target": "self-gemini",
      "value": 0.3333333333333329,
      "lo": 0.24862102140164288,
      "hi": 0.43037681133401817,
      "n": 100,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "arithmetic=inconsistent",
      "target": "self-gemini",
      "value": 0.33333333333333326,
      "lo": 0.15935010395826088,
      "hi": 0.5687559410446896,
      "n": 17,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "arithmetic=inconsistent",
      "target": "self-gemini",
      "value": 0.33333333333333326,
      "lo": 0.1627877475083369,
      "hi": 0.5625053270405489,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "arithmetic=inconsistent",
      "target": "self-gemini",
      "value": 0.3333333333333333,
      "lo": 0.1475552363427626,
      "hi": 0.5908816979031161,
      "n": 14,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "arithmetic=inconsistent",
      "target": "self-gemini",
      "value": 0.33333333333333326,
      "lo": 0.1627877475083369,
      "hi": 0.5625053270405489,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "arithmetic=inconsistent",
      "target": "self-gemini",
      "value": 0.33333333333333326,
      "lo": 0.1627877475083369,
      "hi": 0.5625053270405489,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "arithmetic=inconsistent",
      "target": "self-gemini",
      "value": 0.33333333333333326,
      "lo": 0.1627877475083369,
      "hi": 0.5625053270405489,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "base=sroie",
      "target": "self-gemini",
      "value": 0.33333333333333326,
      "lo": 0.15935010395826088,
      "hi": 0.5687559410446896,
      "n": 17,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "base=sroie",
      "target": "self-gemini",
      "value": 0.3333333333333332,
      "lo": 0.16906782991982014,
      "hi": 0.5513072227644528,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "base=sroie",
      "target": "self-gemini",
      "value": 0.33333333333333326,
      "lo": 0.15568427411889454,
      "hi": 0.5755182860592508,
      "n": 16,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "base=sroie",
      "target": "self-gemini",
      "value": 0.3333333333333332,
      "lo": 0.16906782991982014,
      "hi": 0.5513072227644528,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "base=sroie",
      "target": "self-gemini",
      "value": 0.3333333333333332,
      "lo": 0.16906782991982014,
      "hi": 0.5513072227644528,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "base=sroie",
      "target": "self-gemini",
      "value": 0.3333333333333332,
      "lo": 0.16906782991982014,
      "hi": 0.5513072227644528,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "capture=photo",
      "target": "self-gemini",
      "value": 0.3333333333333333,
      "lo": 0.1475552363427626,
      "hi": 0.5908816979031161,
      "n": 14,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "capture=photo",
      "target": "self-gemini",
      "value": 0.3333333333333332,
      "lo": 0.17194752605083474,
      "hi": 0.5462654802574475,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "capture=photo",
      "target": "self-gemini",
      "value": 0.3333333333333333,
      "lo": 0.1430225574285459,
      "hi": 0.5996759000228535,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.tax",
      "stratum": "capture=photo",
      "target": "self-gemini",
      "value": 0.3333333333333333,
      "lo": 0.13812009109121315,
      "hi": 0.6093779111272004,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "capture=photo",
      "target": "self-gemini",
      "value": 0.3333333333333332,
      "lo": 0.17194752605083474,
      "hi": 0.5462654802574475,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "capture=photo",
      "target": "self-gemini",
      "value": 0.3333333333333332,
      "lo": 0.17194752605083474,
      "hi": 0.5462654802574475,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "line_items_f1",
      "stratum": "capture=photo",
      "target": "self-gemini",
      "value": 0.41346153846153844,
      "lo": 0.3018867924528302,
      "hi": 0.5,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "capture=photo",
      "target": "self-gemini",
      "value": 0.3333333333333332,
      "lo": 0.17194752605083474,
      "hi": 0.5462654802574475,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "capture=scan",
      "target": "self-gemini",
      "value": 0.32921810699588444,
      "lo": 0.2366576417318615,
      "hi": 0.4372439265434072,
      "n": 81,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "capture=scan",
      "target": "self-gemini",
      "value": 0.3296296296296292,
      "lo": 0.2412395665169162,
      "hi": 0.4319681281198166,
      "n": 90,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "capture=scan",
      "target": "self-gemini",
      "value": 0.33333333333333304,
      "lo": 0.236514174104019,
      "hi": 0.44660241902810915,
      "n": 74,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "capture=scan",
      "target": "self-gemini",
      "value": 0.3333333333333329,
      "lo": 0.24823816410787972,
      "hi": 0.43087957324806203,
      "n": 99,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "capture=scan",
      "target": "self-gemini",
      "value": 0.32323232323232276,
      "lo": 0.23920000643397227,
      "hi": 0.42047032106475374,
      "n": 99,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "capture=scan",
      "target": "self-gemini",
      "value": 0.3333333333333329,
      "lo": 0.24823816410787972,
      "hi": 0.43087957324806203,
      "n": 99,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "class=ai_full",
      "target": "self-gemini",
      "value": 0.3333333333333332,
      "lo": 0.16602033991933535,
      "hi": 0.5567060680982032,
      "n": 19,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "class=ai_full",
      "target": "self-gemini",
      "value": 0.3333333333333332,
      "lo": 0.16906782991982014,
      "hi": 0.5513072227644528,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "class=ai_full",
      "target": "self-gemini",
      "value": 0.3333333333333333,
      "lo": 0.13812009109121315,
      "hi": 0.6093779111272004,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.subtotal",
      "stratum": "class=ai_full",
      "target": "self-gemini",
      "value": 0.31249999999999994,
      "lo": 0.14164643854782033,
      "hi": 0.5559564416525933,
      "n": 16,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.tax",
      "stratum": "class=ai_full",
      "target": "self-gemini",
      "value": 0.3333333333333332,
      "lo": 0.16906782991982014,
      "hi": 0.5513072227644528,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "class=ai_full",
      "target": "self-gemini",
      "value": 0.3333333333333332,
      "lo": 0.16906782991982014,
      "hi": 0.5513072227644528,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.receipt_number",
      "stratum": "class=ai_full",
      "target": "self-gemini",
      "value": 0.33333333333333337,
      "lo": 0.132793283187926,
      "hi": 0.6201510391347513,
      "n": 11,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "class=ai_full",
      "target": "self-gemini",
      "value": 0.31666666666666654,
      "lo": 0.15716056329227918,
      "hi": 0.5352519946604211,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "line_items_f1",
      "stratum": "class=ai_full",
      "target": "self-gemini",
      "value": 0.43023255813953487,
      "lo": 0.366656976744186,
      "hi": 0.4876543209876543,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "class=ai_full",
      "target": "self-gemini",
      "value": 0.3333333333333332,
      "lo": 0.16906782991982014,
      "hi": 0.5513072227644528,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "class=ai_inpaint",
      "target": "self-gemini",
      "value": 0.33333333333333337,
      "lo": 0.1269754121301785,
      "hi": 0.632202187824118,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "class=ai_inpaint",
      "target": "self-gemini",
      "value": 0.33333333333333337,
      "lo": 0.1269754121301785,
      "hi": 0.632202187824118,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "class=ai_inpaint",
      "target": "self-gemini",
      "value": 0.33333333333333337,
      "lo": 0.1269754121301785,
      "hi": 0.632202187824118,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "class=ai_inpaint",
      "target": "self-gemini",
      "value": 0.33333333333333337,
      "lo": 0.1269754121301785,
      "hi": 0.632202187824118,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "class=ai_inpaint",
      "target": "self-gemini",
      "value": 0.33333333333333337,
      "lo": 0.1269754121301785,
      "hi": 0.632202187824118,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "class=ai_inpaint",
      "target": "self-gemini",
      "value": 0.33333333333333337,
      "lo": 0.1269754121301785,
      "hi": 0.632202187824118,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "class=genuine",
      "target": "self-gemini",
      "value": 0.32653061224489816,
      "lo": 0.21207497001811879,
      "hi": 0.46620794685429856,
      "n": 49,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "class=genuine",
      "target": "self-gemini",
      "value": 0.32795698924731176,
      "lo": 0.2241533031833493,
      "hi": 0.45183605495905826,
      "n": 62,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "class=genuine",
      "target": "self-gemini",
      "value": 0.3333333333333335,
      "lo": 0.21969931853588529,
      "hi": 0.47031622143029206,
      "n": 51,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "class=genuine",
      "target": "self-gemini",
      "value": 0.3333333333333331,
      "lo": 0.23413420812686653,
      "hi": 0.449873479343222,
      "n": 70,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "class=genuine",
      "target": "self-gemini",
      "value": 0.3238095238095236,
      "lo": 0.22585488569859297,
      "hi": 0.4400960981983579,
      "n": 70,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "class=genuine",
      "target": "self-gemini",
      "value": 0.3333333333333331,
      "lo": 0.23413420812686653,
      "hi": 0.449873479343222,
      "n": 70,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "class=hard_negative",
      "target": "self-gemini",
      "value": 0.33333333333333337,
      "lo": 0.1269754121301785,
      "hi": 0.632202187824118,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "class=hard_negative",
      "target": "self-gemini",
      "value": 0.33333333333333337,
      "lo": 0.1269754121301785,
      "hi": 0.632202187824118,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "class=hard_negative",
      "target": "self-gemini",
      "value": 0.33333333333333337,
      "lo": 0.132793283187926,
      "hi": 0.6201510391347513,
      "n": 11,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "class=hard_negative",
      "target": "self-gemini",
      "value": 0.33333333333333337,
      "lo": 0.132793283187926,
      "hi": 0.6201510391347513,
      "n": 11,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "class=hard_negative",
      "target": "self-gemini",
      "value": 0.33333333333333337,
      "lo": 0.132793283187926,
      "hi": 0.6201510391347513,
      "n": 11,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "class=manual_edit",
      "target": "self-gemini",
      "value": 0.33333333333333337,
      "lo": 0.1269754121301785,
      "hi": 0.632202187824118,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "class=manual_edit",
      "target": "self-gemini",
      "value": 0.33333333333333337,
      "lo": 0.1269754121301785,
      "hi": 0.632202187824118,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "class=manual_edit",
      "target": "self-gemini",
      "value": 0.33333333333333337,
      "lo": 0.1269754121301785,
      "hi": 0.632202187824118,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "class=manual_edit",
      "target": "self-gemini",
      "value": 0.33333333333333337,
      "lo": 0.1269754121301785,
      "hi": 0.632202187824118,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "country=MY",
      "target": "self-gemini",
      "value": 0.33333333333333304,
      "lo": 0.2381881772183168,
      "hi": 0.444317964542494,
      "n": 77,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "country=MY",
      "target": "self-gemini",
      "value": 0.32936507936507897,
      "lo": 0.2382609629131269,
      "hi": 0.43539351583155783,
      "n": 84,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "country=MY",
      "target": "self-gemini",
      "value": 0.3333333333333331,
      "lo": 0.2347460591364293,
      "hi": 0.44902992492920035,
      "n": 71,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "country=MY",
      "target": "self-gemini",
      "value": 0.33333333333333287,
      "lo": 0.24582386759676284,
      "hi": 0.43406529994048143,
      "n": 93,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "country=MY",
      "target": "self-gemini",
      "value": 0.3261648745519709,
      "lo": 0.23943502062452823,
      "hi": 0.42668593906485563,
      "n": 93,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "country=MY",
      "target": "self-gemini",
      "value": 0.33333333333333287,
      "lo": 0.24582386759676284,
      "hi": 0.43406529994048143,
      "n": 93,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "currency=MYR",
      "target": "self-gemini",
      "value": 0.33333333333333304,
      "lo": 0.2381881772183168,
      "hi": 0.444317964542494,
      "n": 77,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "currency=MYR",
      "target": "self-gemini",
      "value": 0.32936507936507897,
      "lo": 0.2382609629131269,
      "hi": 0.43539351583155783,
      "n": 84,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "currency=MYR",
      "target": "self-gemini",
      "value": 0.3333333333333331,
      "lo": 0.2347460591364293,
      "hi": 0.44902992492920035,
      "n": 71,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "currency=MYR",
      "target": "self-gemini",
      "value": 0.33333333333333287,
      "lo": 0.24582386759676284,
      "hi": 0.43406529994048143,
      "n": 93,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "currency=MYR",
      "target": "self-gemini",
      "value": 0.3261648745519709,
      "lo": 0.23943502062452823,
      "hi": 0.42668593906485563,
      "n": 93,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "currency=MYR",
      "target": "self-gemini",
      "value": 0.33333333333333287,
      "lo": 0.24582386759676284,
      "hi": 0.43406529994048143,
      "n": 93,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "dateEra=old",
      "target": "self-gemini",
      "value": 0.333333333333333,
      "lo": 0.24079071397595225,
      "hi": 0.4407928272540806,
      "n": 82,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "dateEra=old",
      "target": "self-gemini",
      "value": 0.329588014981273,
      "lo": 0.24076275412393316,
      "hi": 0.4325153942524667,
      "n": 89,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "dateEra=old",
      "target": "self-gemini",
      "value": 0.33333333333333304,
      "lo": 0.236514174104019,
      "hi": 0.44660241902810915,
      "n": 74,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.tax",
      "stratum": "dateEra=old",
      "target": "self-gemini",
      "value": 0.33333333333333337,
      "lo": 0.1269754121301785,
      "hi": 0.632202187824118,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "dateEra=old",
      "target": "self-gemini",
      "value": 0.33333333333333287,
      "lo": 0.24784994777000835,
      "hi": 0.43139004893721755,
      "n": 98,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "dateEra=old",
      "target": "self-gemini",
      "value": 0.3265306122448975,
      "lo": 0.24176256274018393,
      "hi": 0.4243851889346839,
      "n": 98,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "line_items_f1",
      "stratum": "dateEra=old",
      "target": "self-gemini",
      "value": 0.4523809523809524,
      "lo": 0.38095238095238093,
      "hi": 0.5,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "dateEra=old",
      "target": "self-gemini",
      "value": 0.33333333333333287,
      "lo": 0.24784994777000835,
      "hi": 0.43139004893721755,
      "n": 98,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "dateEra=recent",
      "target": "self-gemini",
      "value": 0.3076923076923077,
      "lo": 0.12680703655710515,
      "hi": 0.5763065681945096,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "dateEra=recent",
      "target": "self-gemini",
      "value": 0.3333333333333332,
      "lo": 0.17726209796920395,
      "hi": 0.5371101081809382,
      "n": 23,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "dateEra=recent",
      "target": "self-gemini",
      "value": 0.3333333333333333,
      "lo": 0.1475552363427626,
      "hi": 0.5908816979031161,
      "n": 14,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.tax",
      "stratum": "dateEra=recent",
      "target": "self-gemini",
      "value": 0.33333333333333337,
      "lo": 0.1269754121301785,
      "hi": 0.632202187824118,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "dateEra=recent",
      "target": "self-gemini",
      "value": 0.3333333333333332,
      "lo": 0.17726209796920395,
      "hi": 0.5371101081809382,
      "n": 23,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "dateEra=recent",
      "target": "self-gemini",
      "value": 0.3188405797101448,
      "lo": 0.16656971007186724,
      "hi": 0.5229652966130698,
      "n": 23,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "line_items_f1",
      "stratum": "dateEra=recent",
      "target": "self-gemini",
      "value": 0.4090909090909091,
      "lo": 0.29590677179962893,
      "hi": 0.4895833333333333,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "dateEra=recent",
      "target": "self-gemini",
      "value": 0.3333333333333332,
      "lo": 0.17726209796920395,
      "hi": 0.5371101081809382,
      "n": 23,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "docType=receipt",
      "target": "self-gemini",
      "value": 0.3298245614035083,
      "lo": 0.24351670349154178,
      "hi": 0.4293601063096113,
      "n": 95,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "docType=receipt",
      "target": "self-gemini",
      "value": 0.33035714285714274,
      "lo": 0.2501477498939,
      "hi": 0.42181770714667,
      "n": 112,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "docType=receipt",
      "target": "self-gemini",
      "value": 0.3333333333333329,
      "lo": 0.2436427243338542,
      "hi": 0.4369662978877758,
      "n": 88,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.subtotal",
      "stratum": "docType=receipt",
      "target": "self-gemini",
      "value": 0.31249999999999994,
      "lo": 0.14164643854782033,
      "hi": 0.5559564416525933,
      "n": 16,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.tax",
      "stratum": "docType=receipt",
      "target": "self-gemini",
      "value": 0.3333333333333332,
      "lo": 0.16906782991982014,
      "hi": 0.5513072227644528,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "docType=receipt",
      "target": "self-gemini",
      "value": 0.3333333333333334,
      "lo": 0.255611110856911,
      "hi": 0.4213124551255819,
      "n": 121,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.receipt_number",
      "stratum": "docType=receipt",
      "target": "self-gemini",
      "value": 0.33333333333333337,
      "lo": 0.132793283187926,
      "hi": 0.6201510391347513,
      "n": 11,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "docType=receipt",
      "target": "self-gemini",
      "value": 0.325068870523416,
      "lo": 0.24811064057894833,
      "hi": 0.412792606361354,
      "n": 121,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "line_items_f1",
      "stratum": "docType=receipt",
      "target": "self-gemini",
      "value": 0.43023255813953487,
      "lo": 0.36902472527472535,
      "hi": 0.4831507490636704,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "docType=receipt",
      "target": "self-gemini",
      "value": 0.3333333333333334,
      "lo": 0.255611110856911,
      "hi": 0.4213124551255819,
      "n": 121,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "lang=en",
      "target": "self-gemini",
      "value": 0.333333333333333,
      "lo": 0.24128544777308758,
      "hi": 0.44012632210756975,
      "n": 83,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "lang=en",
      "target": "self-gemini",
      "value": 0.3296296296296292,
      "lo": 0.2412395665169162,
      "hi": 0.4319681281198166,
      "n": 90,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "lang=en",
      "target": "self-gemini",
      "value": 0.33333333333333304,
      "lo": 0.236514174104019,
      "hi": 0.44660241902810915,
      "n": 74,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.tax",
      "stratum": "lang=en",
      "target": "self-gemini",
      "value": 0.33333333333333337,
      "lo": 0.1269754121301785,
      "hi": 0.632202187824118,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "lang=en",
      "target": "self-gemini",
      "value": 0.3333333333333329,
      "lo": 0.24823816410787972,
      "hi": 0.43087957324806203,
      "n": 99,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "lang=en",
      "target": "self-gemini",
      "value": 0.3265993265993261,
      "lo": 0.24220766476315858,
      "hi": 0.4239451326879726,
      "n": 99,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "line_items_f1",
      "stratum": "lang=en",
      "target": "self-gemini",
      "value": 0.4268292682926829,
      "lo": 0.35,
      "hi": 0.5,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "lang=en",
      "target": "self-gemini",
      "value": 0.3333333333333329,
      "lo": 0.24823816410787972,
      "hi": 0.43087957324806203,
      "n": 99,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "provenance=intact",
      "target": "self-gemini",
      "value": 0.33333333333333326,
      "lo": 0.15935010395826088,
      "hi": 0.5687559410446896,
      "n": 17,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "provenance=intact",
      "target": "self-gemini",
      "value": 0.3333333333333332,
      "lo": 0.16906782991982014,
      "hi": 0.5513072227644528,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "provenance=intact",
      "target": "self-gemini",
      "value": 0.3333333333333333,
      "lo": 0.13812009109121315,
      "hi": 0.6093779111272004,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.tax",
      "stratum": "provenance=intact",
      "target": "self-gemini",
      "value": 0.33333333333333337,
      "lo": 0.1269754121301785,
      "hi": 0.632202187824118,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "provenance=intact",
      "target": "self-gemini",
      "value": 0.3333333333333332,
      "lo": 0.16906782991982014,
      "hi": 0.5513072227644528,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "provenance=intact",
      "target": "self-gemini",
      "value": 0.3333333333333332,
      "lo": 0.16906782991982014,
      "hi": 0.5513072227644528,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "line_items_f1",
      "stratum": "provenance=intact",
      "target": "self-gemini",
      "value": 0.47674418604651164,
      "lo": 0.4268292682926829,
      "hi": 0.5,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "provenance=intact",
      "target": "self-gemini",
      "value": 0.3333333333333332,
      "lo": 0.16906782991982014,
      "hi": 0.5513072227644528,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "provenance=none",
      "target": "self-gemini",
      "value": 0.327683615819209,
      "lo": 0.22169172275612037,
      "hi": 0.4547426930751457,
      "n": 59,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "provenance=none",
      "target": "self-gemini",
      "value": 0.3287037037037035,
      "lo": 0.23130535101308564,
      "hi": 0.4434547729319967,
      "n": 72,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "provenance=none",
      "target": "self-gemini",
      "value": 0.33333333333333326,
      "lo": 0.22729261523522676,
      "hi": 0.45943133547880305,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "provenance=none",
      "target": "self-gemini",
      "value": 0.333333333333333,
      "lo": 0.2402877124166734,
      "hi": 0.4414716493941309,
      "n": 81,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "provenance=none",
      "target": "self-gemini",
      "value": 0.3251028806584359,
      "lo": 0.2330357855180404,
      "hi": 0.4330079892216927,
      "n": 81,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "provenance=none",
      "target": "self-gemini",
      "value": 0.333333333333333,
      "lo": 0.2402877124166734,
      "hi": 0.4414716493941309,
      "n": 81,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "provenance=stripped",
      "target": "self-gemini",
      "value": 0.3333333333333332,
      "lo": 0.16602033991933535,
      "hi": 0.5567060680982032,
      "n": 19,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "provenance=stripped",
      "target": "self-gemini",
      "value": 0.3333333333333332,
      "lo": 0.16906782991982014,
      "hi": 0.5513072227644528,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "provenance=stripped",
      "target": "self-gemini",
      "value": 0.33333333333333326,
      "lo": 0.15568427411889454,
      "hi": 0.5755182860592508,
      "n": 16,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.tax",
      "stratum": "provenance=stripped",
      "target": "self-gemini",
      "value": 0.33333333333333337,
      "lo": 0.1269754121301785,
      "hi": 0.632202187824118,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "provenance=stripped",
      "target": "self-gemini",
      "value": 0.3333333333333332,
      "lo": 0.16906782991982014,
      "hi": 0.5513072227644528,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "provenance=stripped",
      "target": "self-gemini",
      "value": 0.31666666666666654,
      "lo": 0.15716056329227918,
      "hi": 0.5352519946604211,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "line_items_f1",
      "stratum": "provenance=stripped",
      "target": "self-gemini",
      "value": 0.38372093023255816,
      "lo": 0.26744186046511625,
      "hi": 0.47619047619047616,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "provenance=stripped",
      "target": "self-gemini",
      "value": 0.3333333333333332,
      "lo": 0.16906782991982014,
      "hi": 0.5513072227644528,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "source=commons",
      "target": "self-gemini",
      "value": 0.3333333333333333,
      "lo": 0.1430225574285459,
      "hi": 0.5996759000228535,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "source=commons",
      "target": "self-gemini",
      "value": 0.3333333333333333,
      "lo": 0.1430225574285459,
      "hi": 0.5996759000228535,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "source=commons",
      "target": "self-gemini",
      "value": 0.3076923076923077,
      "lo": 0.12680703655710515,
      "hi": 0.5763065681945096,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "source=commons",
      "target": "self-gemini",
      "value": 0.3333333333333333,
      "lo": 0.1430225574285459,
      "hi": 0.5996759000228535,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "source=edited:sharp-patch",
      "target": "self-gemini",
      "value": 0.33333333333333337,
      "lo": 0.1269754121301785,
      "hi": 0.632202187824118,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "source=edited:sharp-patch",
      "target": "self-gemini",
      "value": 0.33333333333333337,
      "lo": 0.1269754121301785,
      "hi": 0.632202187824118,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "source=edited:sharp-patch",
      "target": "self-gemini",
      "value": 0.33333333333333337,
      "lo": 0.1269754121301785,
      "hi": 0.632202187824118,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "source=edited:sharp-patch",
      "target": "self-gemini",
      "value": 0.33333333333333337,
      "lo": 0.1269754121301785,
      "hi": 0.632202187824118,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "source=sroie",
      "target": "self-gemini",
      "value": 0.33333333333333337,
      "lo": 0.22327553774506617,
      "hi": 0.465152762627296,
      "n": 55,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "source=sroie",
      "target": "self-gemini",
      "value": 0.327683615819209,
      "lo": 0.22169172275612037,
      "hi": 0.4547426930751457,
      "n": 59,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "source=sroie",
      "target": "self-gemini",
      "value": 0.3333333333333334,
      "lo": 0.22062697673438705,
      "hi": 0.46897043530690063,
      "n": 52,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "source=sroie",
      "target": "self-gemini",
      "value": 0.33333333333333315,
      "lo": 0.23287456398365092,
      "hi": 0.451615881523039,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "source=sroie",
      "target": "self-gemini",
      "value": 0.32843137254901944,
      "lo": 0.22861902293685715,
      "hi": 0.44659172979061784,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "source=sroie",
      "target": "self-gemini",
      "value": 0.33333333333333315,
      "lo": 0.23287456398365092,
      "hi": 0.451615881523039,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "totals_add_up.recall",
      "stratum": "all",
      "target": "self-gemini",
      "value": 0.12962962962962962,
      "lo": 0.03957024141337484,
      "hi": 0.34996992425081597,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "totals_add_up.precision",
      "stratum": "all",
      "target": "self-gemini",
      "value": 0.8749999999999999,
      "lo": 0.31667744818951943,
      "hi": 0.9906307012744284,
      "n": 8,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "false_alarm_rate",
      "stratum": "all",
      "target": "self-gemini",
      "value": 0.004273504273504273,
      "lo": 0.0003178190465284274,
      "hi": 0.054765821351886415,
      "n": 78,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "checks",
      "metric": "coverage",
      "stratum": "all",
      "target": "self-gemini",
      "value": 0.2341597796143248,
      "lo": 0.16760706704097222,
      "hi": 0.31707267043523474,
      "n": 121,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "false_alarm_rate",
      "stratum": "arithmetic=consistent",
      "target": "self-gemini",
      "value": 0.004273504273504273,
      "lo": 0.0003178190465284274,
      "hi": 0.054765821351886415,
      "n": 78,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "checks",
      "metric": "coverage",
      "stratum": "arithmetic=consistent",
      "target": "self-gemini",
      "value": 0.24999999999999978,
      "lo": 0.17545211362287655,
      "hi": 0.34304463548061576,
      "n": 100,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "totals_add_up.recall",
      "stratum": "arithmetic=inconsistent",
      "target": "self-gemini",
      "value": 0.12962962962962962,
      "lo": 0.03957024141337484,
      "hi": 0.34996992425081597,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "coverage",
      "stratum": "arithmetic=inconsistent",
      "target": "self-gemini",
      "value": 0.14814814814814814,
      "lo": 0.04870716872075084,
      "hi": 0.37135598866023045,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "totals_add_up.recall",
      "stratum": "base=sroie",
      "target": "self-gemini",
      "value": 0.1818181818181818,
      "lo": 0.05136768974608508,
      "hi": 0.47698056196084415,
      "n": 11,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "coverage",
      "stratum": "base=sroie",
      "target": "self-gemini",
      "value": 0.2333333333333333,
      "lo": 0.1011725498384442,
      "hi": 0.4514275344563928,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "coverage",
      "stratum": "capture=photo",
      "target": "self-gemini",
      "value": 0.19047619047619047,
      "lo": 0.07667565052761793,
      "hi": 0.40000564690204926,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "totals_add_up.recall",
      "stratum": "capture=scan",
      "target": "self-gemini",
      "value": 0.17948717948717946,
      "lo": 0.05527092607546122,
      "hi": 0.44991841517722997,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "false_alarm_rate",
      "stratum": "capture=scan",
      "target": "self-gemini",
      "value": 0.004901960784313725,
      "lo": 0.0003645718312947112,
      "hi": 0.06238645746799082,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "checks",
      "metric": "coverage",
      "stratum": "capture=scan",
      "target": "self-gemini",
      "value": 0.2457912457912456,
      "lo": 0.17153340034976716,
      "hi": 0.33904011824363967,
      "n": 99,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "coverage",
      "stratum": "class=ai_full",
      "target": "self-gemini",
      "value": 0.11666666666666665,
      "lo": 0.035533746172033476,
      "hi": 0.3213288750017947,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "coverage",
      "stratum": "class=ai_inpaint",
      "target": "self-gemini",
      "value": 0.16666666666666666,
      "lo": 0.042141900169843094,
      "hi": 0.4762132997387497,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "false_alarm_rate",
      "stratum": "class=genuine",
      "target": "self-gemini",
      "value": 0.004975124378109453,
      "lo": 0.0003700149377350512,
      "hi": 0.06326680963375278,
      "n": 67,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "checks",
      "metric": "coverage",
      "stratum": "class=genuine",
      "target": "self-gemini",
      "value": 0.26666666666666666,
      "lo": 0.17721430257267434,
      "hi": 0.38039645988545023,
      "n": 70,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "false_alarm_rate",
      "stratum": "class=hard_negative",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.2588329669680317,
      "n": 11,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "checks",
      "metric": "coverage",
      "stratum": "class=hard_negative",
      "target": "self-gemini",
      "value": 0.2424242424242424,
      "lo": 0.08110064930098895,
      "hi": 0.5370860306522395,
      "n": 11,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "coverage",
      "stratum": "class=manual_edit",
      "target": "self-gemini",
      "value": 0.3,
      "lo": 0.10779126740630099,
      "hi": 0.6032218525388546,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "totals_add_up.recall",
      "stratum": "country=MY",
      "target": "self-gemini",
      "value": 0.17948717948717946,
      "lo": 0.05527092607546122,
      "hi": 0.44991841517722997,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "false_alarm_rate",
      "stratum": "country=MY",
      "target": "self-gemini",
      "value": 0.00505050505050505,
      "lo": 0.0003756230402271882,
      "hi": 0.06417236269034904,
      "n": 66,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "checks",
      "metric": "coverage",
      "stratum": "country=MY",
      "target": "self-gemini",
      "value": 0.26164874551971307,
      "lo": 0.1830540943751766,
      "hi": 0.35915299468346384,
      "n": 93,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "totals_add_up.recall",
      "stratum": "currency=MYR",
      "target": "self-gemini",
      "value": 0.17948717948717946,
      "lo": 0.05527092607546122,
      "hi": 0.44991841517722997,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "false_alarm_rate",
      "stratum": "currency=MYR",
      "target": "self-gemini",
      "value": 0.00505050505050505,
      "lo": 0.0003756230402271882,
      "hi": 0.06417236269034904,
      "n": 66,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "checks",
      "metric": "coverage",
      "stratum": "currency=MYR",
      "target": "self-gemini",
      "value": 0.26164874551971307,
      "lo": 0.1830540943751766,
      "hi": 0.35915299468346384,
      "n": 93,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "totals_add_up.recall",
      "stratum": "dateEra=old",
      "target": "self-gemini",
      "value": 0.16666666666666666,
      "lo": 0.0512064627237552,
      "hi": 0.425667405768002,
      "n": 14,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "false_alarm_rate",
      "stratum": "dateEra=old",
      "target": "self-gemini",
      "value": 0.00505050505050505,
      "lo": 0.0003756230402271882,
      "hi": 0.06417236269034904,
      "n": 66,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "checks",
      "metric": "coverage",
      "stratum": "dateEra=old",
      "target": "self-gemini",
      "value": 0.2517006802721086,
      "lo": 0.1762597891703736,
      "hi": 0.34587326714855576,
      "n": 98,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "false_alarm_rate",
      "stratum": "dateEra=recent",
      "target": "self-gemini",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "checks",
      "metric": "coverage",
      "stratum": "dateEra=recent",
      "target": "self-gemini",
      "value": 0.15942028985507248,
      "lo": 0.06134924515547946,
      "hi": 0.3549765674122026,
      "n": 23,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "totals_add_up.recall",
      "stratum": "docType=receipt",
      "target": "self-gemini",
      "value": 0.12962962962962962,
      "lo": 0.03957024141337484,
      "hi": 0.34996992425081597,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "false_alarm_rate",
      "stratum": "docType=receipt",
      "target": "self-gemini",
      "value": 0.004273504273504273,
      "lo": 0.0003178190465284274,
      "hi": 0.054765821351886415,
      "n": 78,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "checks",
      "metric": "coverage",
      "stratum": "docType=receipt",
      "target": "self-gemini",
      "value": 0.2341597796143248,
      "lo": 0.16760706704097222,
      "hi": 0.31707267043523474,
      "n": 121,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "totals_add_up.recall",
      "stratum": "lang=en",
      "target": "self-gemini",
      "value": 0.16666666666666666,
      "lo": 0.0512064627237552,
      "hi": 0.425667405768002,
      "n": 14,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "false_alarm_rate",
      "stratum": "lang=en",
      "target": "self-gemini",
      "value": 0.004975124378109453,
      "lo": 0.0003700149377350512,
      "hi": 0.06326680963375278,
      "n": 67,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "checks",
      "metric": "coverage",
      "stratum": "lang=en",
      "target": "self-gemini",
      "value": 0.2525252525252523,
      "lo": 0.17729376939697603,
      "hi": 0.3462446891012412,
      "n": 99,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "coverage",
      "stratum": "provenance=intact",
      "target": "self-gemini",
      "value": 0.18333333333333335,
      "lo": 0.07087125531461269,
      "hi": 0.3978413447855063,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "false_alarm_rate",
      "stratum": "provenance=none",
      "target": "self-gemini",
      "value": 0.004273504273504273,
      "lo": 0.0003178190465284274,
      "hi": 0.054765821351886415,
      "n": 78,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "checks",
      "metric": "coverage",
      "stratum": "provenance=none",
      "target": "self-gemini",
      "value": 0.26337448559670773,
      "lo": 0.17975330932332054,
      "hi": 0.3684235623833778,
      "n": 81,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "totals_add_up.recall",
      "stratum": "provenance=stripped",
      "target": "self-gemini",
      "value": 0.09090909090909091,
      "lo": 0.016232172885884855,
      "hi": 0.37735843645159567,
      "n": 11,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "coverage",
      "stratum": "provenance=stripped",
      "target": "self-gemini",
      "value": 0.16666666666666669,
      "lo": 0.06143139945263601,
      "hi": 0.3793187059159103,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "false_alarm_rate",
      "stratum": "source=commons",
      "target": "self-gemini",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "checks",
      "metric": "coverage",
      "stratum": "source=commons",
      "target": "self-gemini",
      "value": 0.20512820512820512,
      "lo": 0.06814495768843681,
      "hi": 0.47662923626403914,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "coverage",
      "stratum": "source=edited:sharp-patch",
      "target": "self-gemini",
      "value": 0.3,
      "lo": 0.10779126740630099,
      "hi": 0.6032218525388546,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "checks",
      "metric": "false_alarm_rate",
      "stratum": "source=sroie",
      "target": "self-gemini",
      "value": 0.00505050505050505,
      "lo": 0.0003756230402271882,
      "hi": 0.06417236269034904,
      "n": 66,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "checks",
      "metric": "coverage",
      "stratum": "source=sroie",
      "target": "self-gemini",
      "value": 0.27450980392156865,
      "lo": 0.182670880907989,
      "hi": 0.3904632512481214,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "all",
      "target": "self-gemini",
      "value": 0.908179012345679,
      "lo": 0.837337962962963,
      "hi": 0.9663657407407407,
      "n": 121,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "all",
      "target": "self-gemini",
      "value": 0.7,
      "lo": 0.55,
      "hi": 0.875,
      "n": 40,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "all",
      "target": "self-gemini",
      "value": 0.825,
      "lo": 0.675,
      "hi": 0.925,
      "n": 40,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "all",
      "target": "self-gemini",
      "value": 0.825,
      "lo": 0.7,
      "hi": 0.925,
      "n": 40,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "all",
      "target": "self-gemini",
      "value": 0.2333333333333333,
      "lo": 0.12933940758144433,
      "hi": 0.3840587797237752,
      "n": 40,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_5",
      "stratum": "all",
      "target": "self-gemini",
      "value": 0.008230452674897118,
      "lo": 0.001079429439750073,
      "hi": 0.05991424306373648,
      "n": 81,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "all",
      "target": "self-gemini",
      "value": 0.2333333333333333,
      "lo": 0.12933940758144433,
      "hi": 0.3840587797237752,
      "n": 40,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_7",
      "stratum": "all",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.045278085432415204,
      "n": 81,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "base=sroie",
      "target": "self-gemini",
      "value": 0.18333333333333335,
      "lo": 0.07087125531461269,
      "hi": 0.3978413447855063,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "base=sroie",
      "target": "self-gemini",
      "value": 0.18333333333333335,
      "lo": 0.07087125531461269,
      "hi": 0.3978413447855063,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "capture=photo",
      "target": "self-gemini",
      "value": 1,
      "lo": 1,
      "hi": 1,
      "n": 21,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "capture=photo",
      "target": "self-gemini",
      "value": 1,
      "lo": 1,
      "hi": 1,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "capture=photo",
      "target": "self-gemini",
      "value": 1,
      "lo": 1,
      "hi": 1,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "capture=photo",
      "target": "self-gemini",
      "value": 1,
      "lo": 1,
      "hi": 1,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "capture=photo",
      "target": "self-gemini",
      "value": 0.2777777777777778,
      "lo": 0.10461389175889282,
      "hi": 0.5587167778656588,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "capture=photo",
      "target": "self-gemini",
      "value": 0.2777777777777778,
      "lo": 0.10461389175889282,
      "hi": 0.5587167778656588,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "capture=scan",
      "target": "self-gemini",
      "value": 0.8699698189134809,
      "lo": 0.7846516599597587,
      "hi": 0.9519680583501006,
      "n": 99,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "capture=scan",
      "target": "self-gemini",
      "value": 0.6428571428571429,
      "lo": 0.4642857142857143,
      "hi": 0.8571428571428571,
      "n": 28,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "capture=scan",
      "target": "self-gemini",
      "value": 0.75,
      "lo": 0.5357142857142857,
      "hi": 0.8928571428571429,
      "n": 28,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "capture=scan",
      "target": "self-gemini",
      "value": 0.75,
      "lo": 0.6071428571428571,
      "hi": 0.8928571428571429,
      "n": 28,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "capture=scan",
      "target": "self-gemini",
      "value": 0.21428571428571422,
      "lo": 0.1021246033289788,
      "hi": 0.3953858565372186,
      "n": 28,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_5",
      "stratum": "capture=scan",
      "target": "self-gemini",
      "value": 0.009389671361502348,
      "lo": 0.0012316058715024641,
      "hi": 0.06791178384281717,
      "n": 71,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "capture=scan",
      "target": "self-gemini",
      "value": 0.21428571428571422,
      "lo": 0.1021246033289788,
      "hi": 0.3953858565372186,
      "n": 28,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_7",
      "stratum": "capture=scan",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.05132795219689034,
      "n": 71,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "country=MY",
      "target": "self-gemini",
      "value": 0.8520588235294118,
      "lo": 0.7423455882352941,
      "hi": 0.944720588235294,
      "n": 93,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "country=MY",
      "target": "self-gemini",
      "value": 0.56,
      "lo": 0.36,
      "hi": 0.8,
      "n": 25,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "country=MY",
      "target": "self-gemini",
      "value": 0.72,
      "lo": 0.48,
      "hi": 0.88,
      "n": 25,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "country=MY",
      "target": "self-gemini",
      "value": 0.72,
      "lo": 0.56,
      "hi": 0.88,
      "n": 25,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "country=MY",
      "target": "self-gemini",
      "value": 0.18666666666666665,
      "lo": 0.0802002570183862,
      "hi": 0.3766002202367509,
      "n": 25,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_5",
      "stratum": "country=MY",
      "target": "self-gemini",
      "value": 0.00980392156862745,
      "lo": 0.0012859952143584458,
      "hi": 0.07074472686414209,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "country=MY",
      "target": "self-gemini",
      "value": 0.18666666666666665,
      "lo": 0.0802002570183862,
      "hi": 0.3766002202367509,
      "n": 25,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_7",
      "stratum": "country=MY",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.053471336520070537,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "currency=MYR",
      "target": "self-gemini",
      "value": 0.8520588235294118,
      "lo": 0.7544044117647059,
      "hi": 0.9461764705882353,
      "n": 93,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "currency=MYR",
      "target": "self-gemini",
      "value": 0.56,
      "lo": 0.36,
      "hi": 0.8,
      "n": 25,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "currency=MYR",
      "target": "self-gemini",
      "value": 0.72,
      "lo": 0.44,
      "hi": 0.88,
      "n": 25,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "currency=MYR",
      "target": "self-gemini",
      "value": 0.72,
      "lo": 0.52,
      "hi": 0.88,
      "n": 25,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "currency=MYR",
      "target": "self-gemini",
      "value": 0.18666666666666665,
      "lo": 0.0802002570183862,
      "hi": 0.3766002202367509,
      "n": 25,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_5",
      "stratum": "currency=MYR",
      "target": "self-gemini",
      "value": 0.00980392156862745,
      "lo": 0.0012859952143584458,
      "hi": 0.07074472686414209,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "currency=MYR",
      "target": "self-gemini",
      "value": 0.18666666666666665,
      "lo": 0.0802002570183862,
      "hi": 0.3766002202367509,
      "n": 25,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_7",
      "stratum": "currency=MYR",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.053471336520070537,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "dateEra=old",
      "target": "self-gemini",
      "value": 0.8767156862745098,
      "lo": 0.7950796568627451,
      "hi": 0.9534313725490197,
      "n": 98,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "dateEra=old",
      "target": "self-gemini",
      "value": 0.6333333333333333,
      "lo": 0.4666666666666667,
      "hi": 0.8666666666666667,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "dateEra=old",
      "target": "self-gemini",
      "value": 0.7666666666666667,
      "lo": 0.5658333333333347,
      "hi": 0.9,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "dateEra=old",
      "target": "self-gemini",
      "value": 0.7666666666666667,
      "lo": 0.6,
      "hi": 0.9,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "dateEra=old",
      "target": "self-gemini",
      "value": 0.21111111111111105,
      "lo": 0.10255235750147021,
      "hi": 0.38525538077682275,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_5",
      "stratum": "dateEra=old",
      "target": "self-gemini",
      "value": 0.00980392156862745,
      "lo": 0.0012859952143584458,
      "hi": 0.07074472686414209,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "dateEra=old",
      "target": "self-gemini",
      "value": 0.21111111111111105,
      "lo": 0.10255235750147021,
      "hi": 0.38525538077682275,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_7",
      "stratum": "dateEra=old",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.053471336520070537,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "dateEra=recent",
      "target": "self-gemini",
      "value": 1,
      "lo": 1,
      "hi": 1,
      "n": 23,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "dateEra=recent",
      "target": "self-gemini",
      "value": 1,
      "lo": 1,
      "hi": 1,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "dateEra=recent",
      "target": "self-gemini",
      "value": 1,
      "lo": 1,
      "hi": 1,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "dateEra=recent",
      "target": "self-gemini",
      "value": 1,
      "lo": 1,
      "hi": 1,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "dateEra=recent",
      "target": "self-gemini",
      "value": 0.3,
      "lo": 0.10779126740630099,
      "hi": 0.6032218525388546,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_5",
      "stratum": "dateEra=recent",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.22809537235419838,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "dateEra=recent",
      "target": "self-gemini",
      "value": 0.3,
      "lo": 0.10779126740630099,
      "hi": 0.6032218525388546,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_7",
      "stratum": "dateEra=recent",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.22809537235419838,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "docType=receipt",
      "target": "self-gemini",
      "value": 0.908179012345679,
      "lo": 0.8433410493827161,
      "hi": 0.9637422839506172,
      "n": 121,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "docType=receipt",
      "target": "self-gemini",
      "value": 0.7,
      "lo": 0.55,
      "hi": 0.875,
      "n": 40,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "docType=receipt",
      "target": "self-gemini",
      "value": 0.825,
      "lo": 0.65,
      "hi": 0.925,
      "n": 40,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "docType=receipt",
      "target": "self-gemini",
      "value": 0.825,
      "lo": 0.7,
      "hi": 0.925,
      "n": 40,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "docType=receipt",
      "target": "self-gemini",
      "value": 0.2333333333333333,
      "lo": 0.12933940758144433,
      "hi": 0.3840587797237752,
      "n": 40,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_5",
      "stratum": "docType=receipt",
      "target": "self-gemini",
      "value": 0.008230452674897118,
      "lo": 0.001079429439750073,
      "hi": 0.05991424306373648,
      "n": 81,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "docType=receipt",
      "target": "self-gemini",
      "value": 0.2333333333333333,
      "lo": 0.12933940758144433,
      "hi": 0.3840587797237752,
      "n": 40,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_7",
      "stratum": "docType=receipt",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.045278085432415204,
      "n": 81,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "lang=en",
      "target": "self-gemini",
      "value": 0.8770531400966184,
      "lo": 0.7929770531400967,
      "hi": 0.9546014492753622,
      "n": 99,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "lang=en",
      "target": "self-gemini",
      "value": 0.6333333333333333,
      "lo": 0.4666666666666667,
      "hi": 0.8333333333333334,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "lang=en",
      "target": "self-gemini",
      "value": 0.7666666666666667,
      "lo": 0.5333333333333333,
      "hi": 0.9,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "lang=en",
      "target": "self-gemini",
      "value": 0.7666666666666667,
      "lo": 0.6,
      "hi": 0.9,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "lang=en",
      "target": "self-gemini",
      "value": 0.21111111111111105,
      "lo": 0.10255235750147021,
      "hi": 0.38525538077682275,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_5",
      "stratum": "lang=en",
      "target": "self-gemini",
      "value": 0.00966183574879227,
      "lo": 0.0012673394193254278,
      "hi": 0.06977451281438231,
      "n": 69,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "lang=en",
      "target": "self-gemini",
      "value": 0.21111111111111105,
      "lo": 0.10255235750147021,
      "hi": 0.38525538077682275,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_7",
      "stratum": "lang=en",
      "target": "self-gemini",
      "value": 0,
      "lo": 3.469446951953614e-18,
      "hi": 0.05273725818905173,
      "n": 69,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_5",
      "stratum": "source=commons",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.22809537235419838,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_7",
      "stratum": "source=commons",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.22809537235419838,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "source=edited:sharp-patch",
      "target": "self-gemini",
      "value": 0.16666666666666666,
      "lo": 0.042141900169843094,
      "hi": 0.4762132997387497,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "source=edited:sharp-patch",
      "target": "self-gemini",
      "value": 0.16666666666666666,
      "lo": 0.042141900169843094,
      "hi": 0.4762132997387497,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_5",
      "stratum": "source=sroie",
      "target": "self-gemini",
      "value": 0.00980392156862745,
      "lo": 0.0012859952143584458,
      "hi": 0.07074472686414209,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_7",
      "stratum": "source=sroie",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.053471336520070537,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "provenance=intact",
      "target": "self-gemini",
      "value": 0.28333333333333327,
      "lo": 0.1340262550429334,
      "hi": 0.5024613134466216,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "provenance=intact",
      "target": "self-gemini",
      "value": 0.28333333333333327,
      "lo": 0.1340262550429334,
      "hi": 0.5024613134466216,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "provenance=stripped",
      "target": "self-gemini",
      "value": 0.18333333333333335,
      "lo": 0.07087125531461269,
      "hi": 0.3978413447855063,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "provenance=stripped",
      "target": "self-gemini",
      "value": 0.18333333333333335,
      "lo": 0.07087125531461269,
      "hi": 0.3978413447855063,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_5",
      "stratum": "provenance=none",
      "target": "self-gemini",
      "value": 0.008230452674897118,
      "lo": 0.001079429439750073,
      "hi": 0.05991424306373648,
      "n": 81,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_7",
      "stratum": "provenance=none",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.045278085432415204,
      "n": 81,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "class=ai_full",
      "target": "self-gemini",
      "value": 0.9691358024691358,
      "lo": 0.9098688271604939,
      "hi": 1,
      "n": 101,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "class=ai_full",
      "target": "self-gemini",
      "value": 0.85,
      "lo": 0.7,
      "hi": 1,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "class=ai_full",
      "target": "self-gemini",
      "value": 0.95,
      "lo": 0.8,
      "hi": 1,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "class=ai_full",
      "target": "self-gemini",
      "value": 0.95,
      "lo": 0.85,
      "hi": 1,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "class=ai_full",
      "target": "self-gemini",
      "value": 0.28333333333333327,
      "lo": 0.1340262550429334,
      "hi": 0.5024613134466216,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "class=ai_full",
      "target": "self-gemini",
      "value": 0.28333333333333327,
      "lo": 0.1340262550429334,
      "hi": 0.5024613134466216,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "class=ai_inpaint",
      "target": "self-gemini",
      "value": 0.8154320987654321,
      "lo": 0.6518518518518519,
      "hi": 0.9462962962962963,
      "n": 91,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "class=ai_inpaint",
      "target": "self-gemini",
      "value": 0.6,
      "lo": 0.3,
      "hi": 0.9,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "class=ai_inpaint",
      "target": "self-gemini",
      "value": 0.6,
      "lo": 0.3,
      "hi": 0.9,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "class=ai_inpaint",
      "target": "self-gemini",
      "value": 0.6,
      "lo": 0.3,
      "hi": 0.9,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "class=ai_inpaint",
      "target": "self-gemini",
      "value": 0.19999999999999998,
      "lo": 0.056682151454375246,
      "hi": 0.5098375284633582,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "class=ai_inpaint",
      "target": "self-gemini",
      "value": 0.19999999999999998,
      "lo": 0.056682151454375246,
      "hi": 0.5098375284633582,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "class=manual_edit",
      "target": "self-gemini",
      "value": 0.8790123456790123,
      "lo": 0.7253086419753086,
      "hi": 0.9962962962962963,
      "n": 91,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "class=manual_edit",
      "target": "self-gemini",
      "value": 0.5,
      "lo": 0.2,
      "hi": 0.9,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "class=manual_edit",
      "target": "self-gemini",
      "value": 0.8,
      "lo": 0.5,
      "hi": 1,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "class=manual_edit",
      "target": "self-gemini",
      "value": 0.8,
      "lo": 0.5,
      "hi": 1,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "class=manual_edit",
      "target": "self-gemini",
      "value": 0.16666666666666666,
      "lo": 0.042141900169843094,
      "hi": 0.4762132997387497,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "class=manual_edit",
      "target": "self-gemini",
      "value": 0.16666666666666666,
      "lo": 0.042141900169843094,
      "hi": 0.4762132997387497,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "arithmetic=consistent",
      "target": "self-gemini",
      "value": 0.867564534231201,
      "lo": 0.7603815937149271,
      "hi": 0.9517396184062851,
      "n": 103,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "arithmetic=consistent",
      "target": "self-gemini",
      "value": 0.6363636363636364,
      "lo": 0.45454545454545453,
      "hi": 0.8636363636363636,
      "n": 22,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "arithmetic=consistent",
      "target": "self-gemini",
      "value": 0.7727272727272727,
      "lo": 0.5454545454545454,
      "hi": 0.9545454545454546,
      "n": 22,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "arithmetic=consistent",
      "target": "self-gemini",
      "value": 0.7727272727272727,
      "lo": 0.5909090909090909,
      "hi": 0.9545454545454546,
      "n": 22,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "arithmetic=consistent",
      "target": "self-gemini",
      "value": 0.2121212121212121,
      "lo": 0.09158961477654523,
      "hi": 0.4182419796950848,
      "n": 22,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "arithmetic=consistent",
      "target": "self-gemini",
      "value": 0.2121212121212121,
      "lo": 0.09158961477654523,
      "hi": 0.4182419796950848,
      "n": 22,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "arithmetic=inconsistent",
      "target": "self-gemini",
      "value": 0.9578189300411523,
      "lo": 0.8827160493827161,
      "hi": 1,
      "n": 99,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "arithmetic=inconsistent",
      "target": "self-gemini",
      "value": 0.7777777777777778,
      "lo": 0.6111111111111112,
      "hi": 1,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "arithmetic=inconsistent",
      "target": "self-gemini",
      "value": 0.8888888888888888,
      "lo": 0.7222222222222222,
      "hi": 1,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "arithmetic=inconsistent",
      "target": "self-gemini",
      "value": 0.8888888888888888,
      "lo": 0.7222222222222222,
      "hi": 1,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "arithmetic=inconsistent",
      "target": "self-gemini",
      "value": 0.25925925925925924,
      "lo": 0.11300225979498921,
      "hi": 0.49019884788673485,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "arithmetic=inconsistent",
      "target": "self-gemini",
      "value": 0.25925925925925924,
      "lo": 0.11300225979498921,
      "hi": 0.49019884788673485,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "duplicates",
      "metric": "recall",
      "stratum": "all",
      "target": "self-gemini",
      "value": 0.3222222222222222,
      "lo": 0.18366359142696476,
      "hi": 0.5011411705904463,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "duplicates",
      "metric": "recall_exact",
      "stratum": "all",
      "target": "self-gemini",
      "value": 0.02222222222222222,
      "lo": 0.002918587066585515,
      "hi": 0.14999421085520684,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "duplicates",
      "metric": "false_match_rate",
      "stratum": "all",
      "target": "self-gemini",
      "value": 0.01515151515151515,
      "lo": 0.001127599931345813,
      "hi": 0.17332561181034714,
      "n": 22,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "duplicates",
      "metric": "false_match_rate_genuine",
      "stratum": "all",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.061129370526788386,
      "n": 59,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "duplicates",
      "metric": "recall",
      "stratum": "arithmetic=consistent",
      "target": "self-gemini",
      "value": 0.3222222222222222,
      "lo": 0.18366359142696476,
      "hi": 0.5011411705904463,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "duplicates",
      "metric": "recall_exact",
      "stratum": "arithmetic=consistent",
      "target": "self-gemini",
      "value": 0.02222222222222222,
      "lo": 0.002918587066585515,
      "hi": 0.14999421085520684,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "duplicates",
      "metric": "false_match_rate",
      "stratum": "arithmetic=consistent",
      "target": "self-gemini",
      "value": 0.01515151515151515,
      "lo": 0.001127599931345813,
      "hi": 0.17332561181034714,
      "n": 22,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "duplicates",
      "metric": "false_match_rate_genuine",
      "stratum": "arithmetic=consistent",
      "target": "self-gemini",
      "value": 0,
      "lo": 6.938893903907228e-18,
      "hi": 0.06419393671876342,
      "n": 56,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "duplicates",
      "metric": "recall",
      "stratum": "capture=scan",
      "target": "self-gemini",
      "value": 0.31746031746031733,
      "lo": 0.16047502150519727,
      "hi": 0.5309011282610165,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "duplicates",
      "metric": "recall_exact",
      "stratum": "capture=scan",
      "target": "self-gemini",
      "value": 0.031746031746031744,
      "lo": 0.004173443065054516,
      "hi": 0.20413928894392916,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "duplicates",
      "metric": "false_match_rate",
      "stratum": "capture=scan",
      "target": "self-gemini",
      "value": 0.01515151515151515,
      "lo": 0.001127599931345813,
      "hi": 0.17332561181034714,
      "n": 22,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "duplicates",
      "metric": "false_match_rate_genuine",
      "stratum": "capture=scan",
      "target": "self-gemini",
      "value": 0,
      "lo": 6.938893903907228e-18,
      "hi": 0.07269781922049637,
      "n": 49,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "duplicates",
      "metric": "recall",
      "stratum": "class=duplicate",
      "target": "self-gemini",
      "value": 0.3222222222222222,
      "lo": 0.18366359142696476,
      "hi": 0.5011411705904463,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "duplicates",
      "metric": "recall_exact",
      "stratum": "class=duplicate",
      "target": "self-gemini",
      "value": 0.02222222222222222,
      "lo": 0.002918587066585515,
      "hi": 0.14999421085520684,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "duplicates",
      "metric": "false_match_rate",
      "stratum": "class=genuine",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.2588329669680317,
      "n": 11,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "duplicates",
      "metric": "false_match_rate_genuine",
      "stratum": "class=genuine",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.061129370526788386,
      "n": 59,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "duplicates",
      "metric": "false_match_rate",
      "stratum": "class=hard_negative",
      "target": "self-gemini",
      "value": 0.0303030303030303,
      "lo": 0.002257399447080477,
      "hi": 0.3014947816441008,
      "n": 11,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "duplicates",
      "metric": "recall",
      "stratum": "country=MY",
      "target": "self-gemini",
      "value": 0.3222222222222222,
      "lo": 0.18366359142696476,
      "hi": 0.5011411705904463,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "duplicates",
      "metric": "recall_exact",
      "stratum": "country=MY",
      "target": "self-gemini",
      "value": 0.02222222222222222,
      "lo": 0.002918587066585515,
      "hi": 0.14999421085520684,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "duplicates",
      "metric": "false_match_rate",
      "stratum": "country=MY",
      "target": "self-gemini",
      "value": 0.01515151515151515,
      "lo": 0.001127599931345813,
      "hi": 0.17332561181034714,
      "n": 22,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "duplicates",
      "metric": "false_match_rate_genuine",
      "stratum": "country=MY",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.07707356308558039,
      "n": 46,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "duplicates",
      "metric": "recall",
      "stratum": "currency=MYR",
      "target": "self-gemini",
      "value": 0.3222222222222222,
      "lo": 0.18366359142696476,
      "hi": 0.5011411705904463,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "duplicates",
      "metric": "recall_exact",
      "stratum": "currency=MYR",
      "target": "self-gemini",
      "value": 0.02222222222222222,
      "lo": 0.002918587066585515,
      "hi": 0.14999421085520684,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "duplicates",
      "metric": "false_match_rate",
      "stratum": "currency=MYR",
      "target": "self-gemini",
      "value": 0.01515151515151515,
      "lo": 0.001127599931345813,
      "hi": 0.17332561181034714,
      "n": 22,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "duplicates",
      "metric": "false_match_rate_genuine",
      "stratum": "currency=MYR",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.07707356308558039,
      "n": 46,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "duplicates",
      "metric": "recall",
      "stratum": "dateEra=old",
      "target": "self-gemini",
      "value": 0.3222222222222222,
      "lo": 0.18366359142696476,
      "hi": 0.5011411705904463,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "duplicates",
      "metric": "recall_exact",
      "stratum": "dateEra=old",
      "target": "self-gemini",
      "value": 0.02222222222222222,
      "lo": 0.002918587066585515,
      "hi": 0.14999421085520684,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "duplicates",
      "metric": "false_match_rate",
      "stratum": "dateEra=old",
      "target": "self-gemini",
      "value": 0.01515151515151515,
      "lo": 0.001127599931345813,
      "hi": 0.17332561181034714,
      "n": 22,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "duplicates",
      "metric": "false_match_rate_genuine",
      "stratum": "dateEra=old",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.07707356308558039,
      "n": 46,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "duplicates",
      "metric": "false_match_rate_genuine",
      "stratum": "dateEra=recent",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.22809537235419838,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "duplicates",
      "metric": "recall",
      "stratum": "docType=receipt",
      "target": "self-gemini",
      "value": 0.3222222222222222,
      "lo": 0.18366359142696476,
      "hi": 0.5011411705904463,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "duplicates",
      "metric": "recall_exact",
      "stratum": "docType=receipt",
      "target": "self-gemini",
      "value": 0.02222222222222222,
      "lo": 0.002918587066585515,
      "hi": 0.14999421085520684,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "duplicates",
      "metric": "false_match_rate",
      "stratum": "docType=receipt",
      "target": "self-gemini",
      "value": 0.01515151515151515,
      "lo": 0.001127599931345813,
      "hi": 0.17332561181034714,
      "n": 22,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "duplicates",
      "metric": "false_match_rate_genuine",
      "stratum": "docType=receipt",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.061129370526788386,
      "n": 59,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "duplicates",
      "metric": "recall",
      "stratum": "lang=en",
      "target": "self-gemini",
      "value": 0.3222222222222222,
      "lo": 0.18366359142696476,
      "hi": 0.5011411705904463,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "duplicates",
      "metric": "recall_exact",
      "stratum": "lang=en",
      "target": "self-gemini",
      "value": 0.02222222222222222,
      "lo": 0.002918587066585515,
      "hi": 0.14999421085520684,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "duplicates",
      "metric": "false_match_rate",
      "stratum": "lang=en",
      "target": "self-gemini",
      "value": 0.01515151515151515,
      "lo": 0.001127599931345813,
      "hi": 0.17332561181034714,
      "n": 22,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "duplicates",
      "metric": "false_match_rate_genuine",
      "stratum": "lang=en",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.07555760416399632,
      "n": 47,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "duplicates",
      "metric": "recall",
      "stratum": "provenance=none",
      "target": "self-gemini",
      "value": 0.3222222222222222,
      "lo": 0.18366359142696476,
      "hi": 0.5011411705904463,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "duplicates",
      "metric": "recall_exact",
      "stratum": "provenance=none",
      "target": "self-gemini",
      "value": 0.02222222222222222,
      "lo": 0.002918587066585515,
      "hi": 0.14999421085520684,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "duplicates",
      "metric": "false_match_rate",
      "stratum": "provenance=none",
      "target": "self-gemini",
      "value": 0.01515151515151515,
      "lo": 0.001127599931345813,
      "hi": 0.17332561181034714,
      "n": 22,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "duplicates",
      "metric": "false_match_rate_genuine",
      "stratum": "provenance=none",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.061129370526788386,
      "n": 59,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "duplicates",
      "metric": "false_match_rate_genuine",
      "stratum": "source=commons",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.22809537235419838,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "duplicates",
      "metric": "recall",
      "stratum": "source=sroie",
      "target": "self-gemini",
      "value": 0.3222222222222222,
      "lo": 0.18366359142696476,
      "hi": 0.5011411705904463,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "duplicates",
      "metric": "recall_exact",
      "stratum": "source=sroie",
      "target": "self-gemini",
      "value": 0.02222222222222222,
      "lo": 0.002918587066585515,
      "hi": 0.14999421085520684,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "duplicates",
      "metric": "false_match_rate",
      "stratum": "source=sroie",
      "target": "self-gemini",
      "value": 0.01515151515151515,
      "lo": 0.001127599931345813,
      "hi": 0.17332561181034714,
      "n": 22,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "duplicates",
      "metric": "false_match_rate_genuine",
      "stratum": "source=sroie",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.07707356308558039,
      "n": 46,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "decision_consistency",
      "stratum": "all",
      "target": "self-gemini",
      "value": 0.7520661157024794,
      "lo": 0.6681669338901985,
      "hi": 0.82045279689661,
      "n": 121,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "decisions",
      "metric": "total_consistency",
      "stratum": "all",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.03077069794747827,
      "n": 121,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "decisions",
      "metric": "calibration_error",
      "stratum": "all",
      "target": "self-gemini",
      "value": 0.07629752066115707,
      "lo": 0.0466667355371901,
      "hi": 0.12146818181818181,
      "n": 121,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_5",
      "stratum": "all",
      "target": "self-gemini",
      "value": 905.1851851851852,
      "lo": 741.6666666666667,
      "hi": 1121.2962962962965,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_7",
      "stratum": "all",
      "target": "self-gemini",
      "value": 766.6666666666667,
      "lo": 716.6666666666665,
      "hi": 816.6666666666667,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_5",
      "stratum": "assumption=fakes_0_1pct",
      "target": "self-gemini",
      "value": 216.44444444444443,
      "lo": 75,
      "hi": 427.77777777777777,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_7",
      "stratum": "assumption=fakes_0_1pct",
      "target": "self-gemini",
      "value": 76.66666666666667,
      "lo": 71.66666666666666,
      "hi": 81.66666666666669,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_5",
      "stratum": "assumption=fakes_5pct",
      "target": "self-gemini",
      "value": 3966.2551440329225,
      "lo": 3666.6666666666665,
      "hi": 4265.843621399178,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_7",
      "stratum": "assumption=fakes_5pct",
      "target": "self-gemini",
      "value": 3833.333333333334,
      "lo": 3624.9999999999995,
      "hi": 4083.333333333334,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_5",
      "stratum": "assumption=fake_32usd",
      "target": "self-gemini",
      "value": 383.8518518518519,
      "lo": 240,
      "hi": 594.2962962962962,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_7",
      "stratum": "assumption=fake_32usd",
      "target": "self-gemini",
      "value": 245.33333333333334,
      "lo": 231.99999999999997,
      "hi": 261.33333333333337,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_5",
      "stratum": "assumption=fake_182usd",
      "target": "self-gemini",
      "value": 1533.8518518518517,
      "lo": 1349.8333333333335,
      "hi": 1756.7962962962963,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_7",
      "stratum": "assumption=fake_182usd",
      "target": "self-gemini",
      "value": 1395.3333333333333,
      "lo": 1319.5,
      "hi": 1486.3333333333335,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_5",
      "stratum": "assumption=review_8usd",
      "target": "self-gemini",
      "value": 831.8518518518518,
      "lo": 741.6666666666667,
      "hi": 938.7037037037037,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_7",
      "stratum": "assumption=review_8usd",
      "target": "self-gemini",
      "value": 766.6666666666667,
      "lo": 725,
      "hi": 808.3333333333334,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_5",
      "stratum": "assumption=review_73usd",
      "target": "self-gemini",
      "value": 1361.4814814814815,
      "lo": 741.6666666666667,
      "hi": 2262.037037037037,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_7",
      "stratum": "assumption=review_73usd",
      "target": "self-gemini",
      "value": 766.6666666666667,
      "lo": 725,
      "hi": 816.6666666666667,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "approve_all_cost_per_1000",
      "stratum": "all",
      "target": "self-gemini",
      "value": 1000,
      "lo": 1000,
      "hi": 1000,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "approve_all_cost_per_1000",
      "stratum": "assumption=fakes_0_1pct",
      "target": "self-gemini",
      "value": 100,
      "lo": 100,
      "hi": 100,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "approve_all_cost_per_1000",
      "stratum": "assumption=fakes_5pct",
      "target": "self-gemini",
      "value": 5000,
      "lo": 5000,
      "hi": 5000,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "approve_all_cost_per_1000",
      "stratum": "assumption=fake_32usd",
      "target": "self-gemini",
      "value": 320,
      "lo": 320,
      "hi": 320,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "approve_all_cost_per_1000",
      "stratum": "assumption=fake_182usd",
      "target": "self-gemini",
      "value": 1820,
      "lo": 1820,
      "hi": 1820,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "approve_all_cost_per_1000",
      "stratum": "assumption=review_8usd",
      "target": "self-gemini",
      "value": 1000,
      "lo": 1000,
      "hi": 1000,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "approve_all_cost_per_1000",
      "stratum": "assumption=review_73usd",
      "target": "self-gemini",
      "value": 1000,
      "lo": 1000,
      "hi": 1000,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "evidence",
      "metric": "edited_field_named",
      "stratum": "all",
      "target": "self-gemini",
      "value": 0.15,
      "lo": 0.052368745896216595,
      "hi": 0.36041886474075696,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "proof",
      "metric": "fakes_caught_with_proof",
      "stratum": "all",
      "target": "self-gemini",
      "value": 0.1999999999999999,
      "lo": 0.10499989725437694,
      "hi": 0.34757306346399486,
      "n": 40,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "proof",
      "metric": "fakes_caught_with_proof",
      "stratum": "class=ai_full",
      "target": "self-gemini",
      "value": 0.21666666666666665,
      "lo": 0.09076519766966104,
      "hi": 0.43387239189360327,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "proof",
      "metric": "fakes_caught_with_proof",
      "stratum": "class=ai_inpaint",
      "target": "self-gemini",
      "value": 0.19999999999999998,
      "lo": 0.056682151454375246,
      "hi": 0.5098375284633582,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "proof",
      "metric": "fakes_caught_with_proof",
      "stratum": "class=manual_edit",
      "target": "self-gemini",
      "value": 0.16666666666666666,
      "lo": 0.042141900169843094,
      "hi": 0.4762132997387497,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "proof",
      "metric": "genuine_with_proof_rate",
      "stratum": "all",
      "target": "self-gemini",
      "value": 0.008230452674897118,
      "lo": 0.001079429439750073,
      "hi": 0.05991424306373648,
      "n": 81,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "all",
      "target": "self-gemini",
      "value": 4889.5,
      "lo": 4889.5,
      "hi": 4889.5,
      "n": 150,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "all",
      "target": "self-gemini",
      "value": 6382.199999999999,
      "lo": 6382.199999999999,
      "hi": 6382.199999999999,
      "n": 150,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "all",
      "target": "self-gemini",
      "value": 7868.31999999999,
      "lo": 7868.31999999999,
      "hi": 7868.31999999999,
      "n": 150,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "all",
      "target": "self-gemini",
      "value": 0.01633333333333333,
      "lo": 0.01633333333333333,
      "hi": 0.01633333333333333,
      "n": 150,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "all",
      "target": "self-gemini",
      "value": 0.004410524999999998,
      "lo": 0.004410524999999998,
      "hi": 0.004410524999999998,
      "n": 150,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "all",
      "target": "self-gemini",
      "value": 0.6688741721854304,
      "lo": 0.6242755037527579,
      "hi": 0.7106328042518599,
      "n": 453,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "all",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.008408735123582253,
      "n": 453,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "all",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.008408735123582253,
      "n": 453,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "arithmetic=consistent",
      "target": "self-gemini",
      "value": 4840,
      "lo": 4840,
      "hi": 4840,
      "n": 129,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "arithmetic=consistent",
      "target": "self-gemini",
      "value": 6194.8,
      "lo": 6194.8,
      "hi": 6194.8,
      "n": 129,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "arithmetic=consistent",
      "target": "self-gemini",
      "value": 8106.039999999999,
      "lo": 8106.039999999999,
      "hi": 8106.039999999999,
      "n": 129,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "arithmetic=consistent",
      "target": "self-gemini",
      "value": 0.01633333333333333,
      "lo": 0.01633333333333333,
      "hi": 0.01633333333333333,
      "n": 129,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "arithmetic=consistent",
      "target": "self-gemini",
      "value": 0.004379232558139535,
      "lo": 0.004379232558139535,
      "hi": 0.004379232558139535,
      "n": 129,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "arithmetic=consistent",
      "target": "self-gemini",
      "value": 0.6692307692307692,
      "lo": 0.6210845365546219,
      "hi": 0.7140757088782902,
      "n": 390,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "arithmetic=consistent",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.009753820311850517,
      "n": 390,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "arithmetic=consistent",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.009753820311850517,
      "n": 390,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "arithmetic=inconsistent",
      "target": "self-gemini",
      "value": 5330.5,
      "lo": 5330.5,
      "hi": 5330.5,
      "n": 18,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "arithmetic=inconsistent",
      "target": "self-gemini",
      "value": 6554.099999999999,
      "lo": 6554.099999999999,
      "hi": 6554.099999999999,
      "n": 18,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "arithmetic=inconsistent",
      "target": "self-gemini",
      "value": 6726.82,
      "lo": 6726.82,
      "hi": 6726.82,
      "n": 18,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "arithmetic=inconsistent",
      "target": "self-gemini",
      "value": 0.01633333333333333,
      "lo": 0.01633333333333333,
      "hi": 0.01633333333333333,
      "n": 18,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "arithmetic=inconsistent",
      "target": "self-gemini",
      "value": 0.004677291666666667,
      "lo": 0.004677291666666667,
      "hi": 0.004677291666666667,
      "n": 18,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "arithmetic=inconsistent",
      "target": "self-gemini",
      "value": 0.6666666666666666,
      "lo": 0.5336096360112935,
      "hi": 0.7775858346250742,
      "n": 54,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "arithmetic=inconsistent",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.0664135880908964,
      "n": 54,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "arithmetic=inconsistent",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.0664135880908964,
      "n": 54,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "arithmeticError=totals",
      "target": "self-gemini",
      "value": 0.6666666666666666,
      "lo": 0.4537345197425524,
      "hi": 0.8280524739491653,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "arithmeticError=totals",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.154639018924847,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "arithmeticError=totals",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.154639018924847,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "base=sroie",
      "target": "self-gemini",
      "value": 5334,
      "lo": 5334,
      "hi": 5334,
      "n": 20,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "base=sroie",
      "target": "self-gemini",
      "value": 6428.95,
      "lo": 6428.95,
      "hi": 6428.95,
      "n": 20,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "base=sroie",
      "target": "self-gemini",
      "value": 6701.79,
      "lo": 6701.79,
      "hi": 6701.79,
      "n": 20,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "base=sroie",
      "target": "self-gemini",
      "value": 0.01633333333333333,
      "lo": 0.01633333333333333,
      "hi": 0.01633333333333333,
      "n": 20,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "base=sroie",
      "target": "self-gemini",
      "value": 0.0045520125,
      "lo": 0.0045520125,
      "hi": 0.0045520125,
      "n": 20,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "base=sroie",
      "target": "self-gemini",
      "value": 0.6666666666666666,
      "lo": 0.5405686645211968,
      "hi": 0.7727073847647731,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "base=sroie",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.06017185214208986,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "base=sroie",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.06017185214208986,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "capture=photo",
      "target": "self-gemini",
      "value": 4913,
      "lo": 4913,
      "hi": 4913,
      "n": 25,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "capture=photo",
      "target": "self-gemini",
      "value": 5984.799999999999,
      "lo": 5984.799999999999,
      "hi": 5984.799999999999,
      "n": 25,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "capture=photo",
      "target": "self-gemini",
      "value": 6408.959999999999,
      "lo": 6408.959999999999,
      "hi": 6408.959999999999,
      "n": 25,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "capture=photo",
      "target": "self-gemini",
      "value": 0.01633333333333333,
      "lo": 0.01633333333333333,
      "hi": 0.01633333333333333,
      "n": 25,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "capture=photo",
      "target": "self-gemini",
      "value": 0.00447546,
      "lo": 0.00447546,
      "hi": 0.00447546,
      "n": 25,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "capture=photo",
      "target": "self-gemini",
      "value": 0.6666666666666666,
      "lo": 0.5541742611414865,
      "hi": 0.7629177913642448,
      "n": 75,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "capture=photo",
      "target": "self-gemini",
      "value": 0,
      "lo": 3.469446951953614e-18,
      "hi": 0.04872384248280587,
      "n": 75,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "capture=photo",
      "target": "self-gemini",
      "value": 0,
      "lo": 3.469446951953614e-18,
      "hi": 0.04872384248280587,
      "n": 75,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "capture=scan",
      "target": "self-gemini",
      "value": 4840,
      "lo": 4840,
      "hi": 4840,
      "n": 119,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "capture=scan",
      "target": "self-gemini",
      "value": 6190.1,
      "lo": 6190.1,
      "hi": 6190.1,
      "n": 119,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "capture=scan",
      "target": "self-gemini",
      "value": 7197.219999999997,
      "lo": 7197.219999999997,
      "hi": 7197.219999999997,
      "n": 119,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "capture=scan",
      "target": "self-gemini",
      "value": 0.01633333333333333,
      "lo": 0.01633333333333333,
      "hi": 0.01633333333333333,
      "n": 119,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "capture=scan",
      "target": "self-gemini",
      "value": 0.004360317226890757,
      "lo": 0.004360317226890757,
      "hi": 0.004360317226890757,
      "n": 119,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "capture=scan",
      "target": "self-gemini",
      "value": 0.6694444444444444,
      "lo": 0.6192862564240844,
      "hi": 0.7160246242969694,
      "n": 360,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "capture=scan",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.010558056888693507,
      "n": 360,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "capture=scan",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.010558056888693507,
      "n": 360,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "capture=screenshot",
      "target": "self-gemini",
      "value": 0.6666666666666666,
      "lo": 0.41713547738519774,
      "hi": 0.8482367556028526,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "capture=screenshot",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "capture=screenshot",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "class=ai_full",
      "target": "self-gemini",
      "value": 4882.5,
      "lo": 4882.5,
      "hi": 4882.5,
      "n": 20,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "class=ai_full",
      "target": "self-gemini",
      "value": 6092.3,
      "lo": 6092.3,
      "hi": 6092.3,
      "n": 20,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "class=ai_full",
      "target": "self-gemini",
      "value": 6431.259999999999,
      "lo": 6431.259999999999,
      "hi": 6431.259999999999,
      "n": 20,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "class=ai_full",
      "target": "self-gemini",
      "value": 0.01633333333333333,
      "lo": 0.01633333333333333,
      "hi": 0.01633333333333333,
      "n": 20,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "class=ai_full",
      "target": "self-gemini",
      "value": 0.00451875,
      "lo": 0.00451875,
      "hi": 0.00451875,
      "n": 20,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "class=ai_full",
      "target": "self-gemini",
      "value": 0.6666666666666666,
      "lo": 0.5405686645211968,
      "hi": 0.7727073847647731,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "class=ai_full",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.06017185214208986,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "class=ai_full",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.06017185214208986,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "class=ai_inpaint",
      "target": "self-gemini",
      "value": 5363.5,
      "lo": 5363.5,
      "hi": 5363.5,
      "n": 10,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "class=ai_inpaint",
      "target": "self-gemini",
      "value": 6608.45,
      "lo": 6608.45,
      "hi": 6608.45,
      "n": 10,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "class=ai_inpaint",
      "target": "self-gemini",
      "value": 6737.6900000000005,
      "lo": 6737.6900000000005,
      "hi": 6737.6900000000005,
      "n": 10,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "class=ai_inpaint",
      "target": "self-gemini",
      "value": 0.01633333333333333,
      "lo": 0.01633333333333333,
      "hi": 0.01633333333333333,
      "n": 10,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "class=ai_inpaint",
      "target": "self-gemini",
      "value": 0.004353825,
      "lo": 0.004353825,
      "hi": 0.004353825,
      "n": 10,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "class=ai_inpaint",
      "target": "self-gemini",
      "value": 0.6666666666666666,
      "lo": 0.4878005164454384,
      "hi": 0.8076950191632386,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "class=ai_inpaint",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "class=ai_inpaint",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "class=duplicate",
      "target": "self-gemini",
      "value": 5010,
      "lo": 5010,
      "hi": 5010,
      "n": 29,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "class=duplicate",
      "target": "self-gemini",
      "value": 7076.5999999999985,
      "lo": 7076.5999999999985,
      "hi": 7076.5999999999985,
      "n": 29,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "class=duplicate",
      "target": "self-gemini",
      "value": 9153.639999999998,
      "lo": 9153.639999999998,
      "hi": 9153.639999999998,
      "n": 29,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "class=duplicate",
      "target": "self-gemini",
      "value": 0.01633333333333333,
      "lo": 0.01633333333333333,
      "hi": 0.01633333333333333,
      "n": 29,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "class=duplicate",
      "target": "self-gemini",
      "value": 0.004482801724137932,
      "lo": 0.004482801724137932,
      "hi": 0.004482801724137932,
      "n": 29,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "class=duplicate",
      "target": "self-gemini",
      "value": 0.6777777777777778,
      "lo": 0.5756683783279567,
      "hi": 0.7653322881380605,
      "n": 90,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "class=duplicate",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.04093562556432678,
      "n": 90,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "class=duplicate",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.04093562556432678,
      "n": 90,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "class=genuine",
      "target": "self-gemini",
      "value": 4713.5,
      "lo": 4713.5,
      "hi": 4713.5,
      "n": 70,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "class=genuine",
      "target": "self-gemini",
      "value": 5587.749999999999,
      "lo": 5587.749999999999,
      "hi": 5587.749999999999,
      "n": 70,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "class=genuine",
      "target": "self-gemini",
      "value": 7452.170000000003,
      "lo": 7452.170000000003,
      "hi": 7452.170000000003,
      "n": 70,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "class=genuine",
      "target": "self-gemini",
      "value": 0.01633333333333333,
      "lo": 0.01633333333333333,
      "hi": 0.01633333333333333,
      "n": 70,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "class=genuine",
      "target": "self-gemini",
      "value": 0.004325025,
      "lo": 0.004325025,
      "hi": 0.004325025,
      "n": 70,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "class=genuine",
      "target": "self-gemini",
      "value": 0.6666666666666666,
      "lo": 0.6004193945242017,
      "hi": 0.7269259218893706,
      "n": 210,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "class=genuine",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.017964050759283236,
      "n": 210,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "class=genuine",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.017964050759283236,
      "n": 210,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "class=hard_negative",
      "target": "self-gemini",
      "value": 5077,
      "lo": 5077,
      "hi": 5077,
      "n": 11,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "class=hard_negative",
      "target": "self-gemini",
      "value": 5791.5,
      "lo": 5791.5,
      "hi": 5791.5,
      "n": 11,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "class=hard_negative",
      "target": "self-gemini",
      "value": 5957.5,
      "lo": 5957.5,
      "hi": 5957.5,
      "n": 11,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "class=hard_negative",
      "target": "self-gemini",
      "value": 0.01633333333333333,
      "lo": 0.01633333333333333,
      "hi": 0.01633333333333333,
      "n": 11,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "class=hard_negative",
      "target": "self-gemini",
      "value": 0.004310045454545455,
      "lo": 0.004310045454545455,
      "hi": 0.004310045454545455,
      "n": 11,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "class=hard_negative",
      "target": "self-gemini",
      "value": 0.6666666666666666,
      "lo": 0.4960789442780561,
      "hi": 0.8024977232077931,
      "n": 33,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "class=hard_negative",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.10426999754245206,
      "n": 33,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "class=hard_negative",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.10426999754245206,
      "n": 33,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "class=manual_edit",
      "target": "self-gemini",
      "value": 5308.5,
      "lo": 5308.5,
      "hi": 5308.5,
      "n": 10,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "class=manual_edit",
      "target": "self-gemini",
      "value": 5896.299999999999,
      "lo": 5896.299999999999,
      "hi": 5896.299999999999,
      "n": 10,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "class=manual_edit",
      "target": "self-gemini",
      "value": 6006.46,
      "lo": 6006.46,
      "hi": 6006.46,
      "n": 10,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "class=manual_edit",
      "target": "self-gemini",
      "value": 0.01633333333333333,
      "lo": 0.01633333333333333,
      "hi": 0.01633333333333333,
      "n": 10,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "class=manual_edit",
      "target": "self-gemini",
      "value": 0.004750200000000001,
      "lo": 0.004750200000000001,
      "hi": 0.004750200000000001,
      "n": 10,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "class=manual_edit",
      "target": "self-gemini",
      "value": 0.6666666666666666,
      "lo": 0.4878005164454384,
      "hi": 0.8076950191632386,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "class=manual_edit",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "class=manual_edit",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "country=JP",
      "target": "self-gemini",
      "value": 0.6666666666666666,
      "lo": 0.39062208887279953,
      "hi": 0.8618799089087867,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "country=JP",
      "target": "self-gemini",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "country=JP",
      "target": "self-gemini",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "country=MY",
      "target": "self-gemini",
      "value": 4889.5,
      "lo": 4889.5,
      "hi": 4889.5,
      "n": 122,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "country=MY",
      "target": "self-gemini",
      "value": 6339.649999999999,
      "lo": 6339.649999999999,
      "hi": 6339.649999999999,
      "n": 122,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "country=MY",
      "target": "self-gemini",
      "value": 8185.279999999991,
      "lo": 8185.279999999991,
      "hi": 8185.279999999991,
      "n": 122,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "country=MY",
      "target": "self-gemini",
      "value": 0.01633333333333333,
      "lo": 0.01633333333333333,
      "hi": 0.01633333333333333,
      "n": 122,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "country=MY",
      "target": "self-gemini",
      "value": 0.004397803278688524,
      "lo": 0.004397803278688524,
      "hi": 0.004397803278688524,
      "n": 122,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "country=MY",
      "target": "self-gemini",
      "value": 0.6693766937669376,
      "lo": 0.6198480847564948,
      "hi": 0.7154150600541113,
      "n": 369,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "country=MY",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.010303196519090837,
      "n": 369,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "country=MY",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.010303196519090837,
      "n": 369,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "currency=JPY",
      "target": "self-gemini",
      "value": 0.6666666666666666,
      "lo": 0.39062208887279953,
      "hi": 0.8618799089087867,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "currency=JPY",
      "target": "self-gemini",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "currency=JPY",
      "target": "self-gemini",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "currency=MYR",
      "target": "self-gemini",
      "value": 4889.5,
      "lo": 4889.5,
      "hi": 4889.5,
      "n": 122,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "currency=MYR",
      "target": "self-gemini",
      "value": 6339.649999999999,
      "lo": 6339.649999999999,
      "hi": 6339.649999999999,
      "n": 122,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "currency=MYR",
      "target": "self-gemini",
      "value": 8185.279999999991,
      "lo": 8185.279999999991,
      "hi": 8185.279999999991,
      "n": 122,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "currency=MYR",
      "target": "self-gemini",
      "value": 0.01633333333333333,
      "lo": 0.01633333333333333,
      "hi": 0.01633333333333333,
      "n": 122,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "currency=MYR",
      "target": "self-gemini",
      "value": 0.004397803278688524,
      "lo": 0.004397803278688524,
      "hi": 0.004397803278688524,
      "n": 122,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "currency=MYR",
      "target": "self-gemini",
      "value": 0.6693766937669376,
      "lo": 0.6198480847564948,
      "hi": 0.7154150600541113,
      "n": 369,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "currency=MYR",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.010303196519090837,
      "n": 369,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "currency=MYR",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.010303196519090837,
      "n": 369,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "dateEra=old",
      "target": "self-gemini",
      "value": 4877,
      "lo": 4877,
      "hi": 4877,
      "n": 127,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "dateEra=old",
      "target": "self-gemini",
      "value": 6302.899999999999,
      "lo": 6302.899999999999,
      "hi": 6302.899999999999,
      "n": 127,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "dateEra=old",
      "target": "self-gemini",
      "value": 8128.679999999994,
      "lo": 8128.679999999994,
      "hi": 8128.679999999994,
      "n": 127,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "dateEra=old",
      "target": "self-gemini",
      "value": 0.01633333333333333,
      "lo": 0.01633333333333333,
      "hi": 0.01633333333333333,
      "n": 127,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "dateEra=old",
      "target": "self-gemini",
      "value": 0.00440002559055118,
      "lo": 0.00440002559055118,
      "hi": 0.00440002559055118,
      "n": 127,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "dateEra=old",
      "target": "self-gemini",
      "value": 0.6692708333333334,
      "lo": 0.6207413618269642,
      "hi": 0.7144471463724682,
      "n": 384,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "dateEra=old",
      "target": "self-gemini",
      "value": 0,
      "lo": 8.673617379884035e-19,
      "hi": 0.009904714241677031,
      "n": 384,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "dateEra=old",
      "target": "self-gemini",
      "value": 0,
      "lo": 8.673617379884035e-19,
      "hi": 0.009904714241677031,
      "n": 384,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "dateEra=recent",
      "target": "self-gemini",
      "value": 4913,
      "lo": 4913,
      "hi": 4913,
      "n": 23,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "dateEra=recent",
      "target": "self-gemini",
      "value": 6471.4,
      "lo": 6471.4,
      "hi": 6471.4,
      "n": 23,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "dateEra=recent",
      "target": "self-gemini",
      "value": 6906.000000000001,
      "lo": 6906.000000000001,
      "hi": 6906.000000000001,
      "n": 23,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "dateEra=recent",
      "target": "self-gemini",
      "value": 0.01633333333333333,
      "lo": 0.01633333333333333,
      "hi": 0.01633333333333333,
      "n": 23,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "dateEra=recent",
      "target": "self-gemini",
      "value": 0.004468499999999999,
      "lo": 0.004468499999999999,
      "hi": 0.004468499999999999,
      "n": 23,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "dateEra=recent",
      "target": "self-gemini",
      "value": 0.6666666666666666,
      "lo": 0.5492647542871832,
      "hi": 0.7664894929831328,
      "n": 69,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "dateEra=recent",
      "target": "self-gemini",
      "value": 0,
      "lo": 3.469446951953614e-18,
      "hi": 0.05273725818905173,
      "n": 69,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "dateEra=recent",
      "target": "self-gemini",
      "value": 0,
      "lo": 3.469446951953614e-18,
      "hi": 0.05273725818905173,
      "n": 69,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "docType=receipt",
      "target": "self-gemini",
      "value": 4889.5,
      "lo": 4889.5,
      "hi": 4889.5,
      "n": 150,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "docType=receipt",
      "target": "self-gemini",
      "value": 6382.199999999999,
      "lo": 6382.199999999999,
      "hi": 6382.199999999999,
      "n": 150,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "docType=receipt",
      "target": "self-gemini",
      "value": 7868.31999999999,
      "lo": 7868.31999999999,
      "hi": 7868.31999999999,
      "n": 150,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "docType=receipt",
      "target": "self-gemini",
      "value": 0.01633333333333333,
      "lo": 0.01633333333333333,
      "hi": 0.01633333333333333,
      "n": 150,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "docType=receipt",
      "target": "self-gemini",
      "value": 0.004410524999999998,
      "lo": 0.004410524999999998,
      "hi": 0.004410524999999998,
      "n": 150,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "docType=receipt",
      "target": "self-gemini",
      "value": 0.6688741721854304,
      "lo": 0.6242755037527579,
      "hi": 0.7106328042518599,
      "n": 453,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "docType=receipt",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.008408735123582253,
      "n": 453,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "docType=receipt",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.008408735123582253,
      "n": 453,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "editedField=date",
      "target": "self-gemini",
      "value": 0.6666666666666666,
      "lo": 0.39062208887279953,
      "hi": 0.8618799089087867,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "editedField=date",
      "target": "self-gemini",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "editedField=date",
      "target": "self-gemini",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "editedField=items+total",
      "target": "self-gemini",
      "value": 0.6666666666666666,
      "lo": 0.39062208887279953,
      "hi": 0.8618799089087867,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "editedField=items+total",
      "target": "self-gemini",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "editedField=items+total",
      "target": "self-gemini",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "editedField=total",
      "target": "self-gemini",
      "value": 0.6666666666666666,
      "lo": 0.47824760507320074,
      "hi": 0.8135673850128435,
      "n": 27,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "editedField=total",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.12455502974186705,
      "n": 27,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "editedField=total",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.12455502974186705,
      "n": 27,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "lang=en",
      "target": "self-gemini",
      "value": 4910.5,
      "lo": 4910.5,
      "hi": 4910.5,
      "n": 128,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "lang=en",
      "target": "self-gemini",
      "value": 6388.599999999999,
      "lo": 6388.599999999999,
      "hi": 6388.599999999999,
      "n": 128,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "lang=en",
      "target": "self-gemini",
      "value": 8117.360000000004,
      "lo": 8117.360000000004,
      "hi": 8117.360000000004,
      "n": 128,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "lang=en",
      "target": "self-gemini",
      "value": 0.01633333333333333,
      "lo": 0.01633333333333333,
      "hi": 0.01633333333333333,
      "n": 128,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "lang=en",
      "target": "self-gemini",
      "value": 0.004403671874999999,
      "lo": 0.004403671874999999,
      "hi": 0.004403671874999999,
      "n": 128,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "lang=en",
      "target": "self-gemini",
      "value": 0.6692506459948321,
      "lo": 0.620913920630256,
      "hi": 0.7142603477169992,
      "n": 387,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "lang=en",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.009828688164979105,
      "n": 387,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "lang=en",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.009828688164979105,
      "n": 387,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "lang=ja",
      "target": "self-gemini",
      "value": 0.6666666666666666,
      "lo": 0.39062208887279953,
      "hi": 0.8618799089087867,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "lang=ja",
      "target": "self-gemini",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "lang=ja",
      "target": "self-gemini",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "lang=zh",
      "target": "self-gemini",
      "value": 0.6666666666666666,
      "lo": 0.43749467295945105,
      "hi": 0.837212252491663,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "lang=zh",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.17587922364665765,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "lang=zh",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.17587922364665765,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "prompt=explicit_amounts",
      "target": "self-gemini",
      "value": 0.6666666666666666,
      "lo": 0.4670631683813174,
      "hi": 0.8202780967270225,
      "n": 24,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "prompt=explicit_amounts",
      "target": "self-gemini",
      "value": 0,
      "lo": 1.3877787807814457e-17,
      "hi": 0.13797620467498017,
      "n": 24,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "prompt=explicit_amounts",
      "target": "self-gemini",
      "value": 0,
      "lo": 1.3877787807814457e-17,
      "hi": 0.13797620467498017,
      "n": 24,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "prompt=free",
      "target": "self-gemini",
      "value": 0.6666666666666666,
      "lo": 0.41713547738519774,
      "hi": 0.8482367556028526,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "prompt=free",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "prompt=free",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "prompt=target_total",
      "target": "self-gemini",
      "value": 0.6666666666666666,
      "lo": 0.4537345197425524,
      "hi": 0.8280524739491653,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "prompt=target_total",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.154639018924847,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "prompt=target_total",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.154639018924847,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "provenance=intact",
      "target": "self-gemini",
      "value": 5085.5,
      "lo": 5085.5,
      "hi": 5085.5,
      "n": 20,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "provenance=intact",
      "target": "self-gemini",
      "value": 6428.95,
      "lo": 6428.95,
      "hi": 6428.95,
      "n": 20,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "provenance=intact",
      "target": "self-gemini",
      "value": 6701.79,
      "lo": 6701.79,
      "hi": 6701.79,
      "n": 20,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "provenance=intact",
      "target": "self-gemini",
      "value": 0.01633333333333333,
      "lo": 0.01633333333333333,
      "hi": 0.01633333333333333,
      "n": 20,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "provenance=intact",
      "target": "self-gemini",
      "value": 0.0045865125,
      "lo": 0.0045865125,
      "hi": 0.0045865125,
      "n": 20,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "provenance=intact",
      "target": "self-gemini",
      "value": 0.6666666666666666,
      "lo": 0.5405686645211968,
      "hi": 0.7727073847647731,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "provenance=intact",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.06017185214208986,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "provenance=intact",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.06017185214208986,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "provenance=none",
      "target": "self-gemini",
      "value": 4755,
      "lo": 4755,
      "hi": 4755,
      "n": 110,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "provenance=none",
      "target": "self-gemini",
      "value": 6280.849999999999,
      "lo": 6280.849999999999,
      "hi": 6280.849999999999,
      "n": 110,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "provenance=none",
      "target": "self-gemini",
      "value": 8321.119999999995,
      "lo": 8321.119999999995,
      "hi": 8321.119999999995,
      "n": 110,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "provenance=none",
      "target": "self-gemini",
      "value": 0.01633333333333333,
      "lo": 0.01633333333333333,
      "hi": 0.01633333333333333,
      "n": 110,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "provenance=none",
      "target": "self-gemini",
      "value": 0.004365122727272727,
      "lo": 0.004365122727272727,
      "hi": 0.004365122727272727,
      "n": 110,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "provenance=none",
      "target": "self-gemini",
      "value": 0.6696696696696697,
      "lo": 0.617470135633495,
      "hi": 0.7179992581745626,
      "n": 333,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "provenance=none",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.011404352760326315,
      "n": 333,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "provenance=none",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.011404352760326315,
      "n": 333,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "provenance=stripped",
      "target": "self-gemini",
      "value": 5334,
      "lo": 5334,
      "hi": 5334,
      "n": 20,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "provenance=stripped",
      "target": "self-gemini",
      "value": 6205.35,
      "lo": 6205.35,
      "hi": 6205.35,
      "n": 20,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "provenance=stripped",
      "target": "self-gemini",
      "value": 6453.87,
      "lo": 6453.87,
      "hi": 6453.87,
      "n": 20,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "provenance=stripped",
      "target": "self-gemini",
      "value": 0.01633333333333333,
      "lo": 0.01633333333333333,
      "hi": 0.01633333333333333,
      "n": 20,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "provenance=stripped",
      "target": "self-gemini",
      "value": 0.00448425,
      "lo": 0.00448425,
      "hi": 0.00448425,
      "n": 20,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "provenance=stripped",
      "target": "self-gemini",
      "value": 0.6666666666666666,
      "lo": 0.5405686645211968,
      "hi": 0.7727073847647731,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "provenance=stripped",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.06017185214208986,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "provenance=stripped",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.06017185214208986,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "source=commons",
      "target": "self-gemini",
      "value": 4811,
      "lo": 4811,
      "hi": 4811,
      "n": 13,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "source=commons",
      "target": "self-gemini",
      "value": 5990.599999999998,
      "lo": 5990.599999999998,
      "hi": 5990.599999999998,
      "n": 13,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "source=commons",
      "target": "self-gemini",
      "value": 6810.919999999998,
      "lo": 6810.919999999998,
      "hi": 6810.919999999998,
      "n": 13,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "source=commons",
      "target": "self-gemini",
      "value": 0.01633333333333333,
      "lo": 0.01633333333333333,
      "hi": 0.01633333333333333,
      "n": 13,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "source=commons",
      "target": "self-gemini",
      "value": 0.00439725,
      "lo": 0.00439725,
      "hi": 0.00439725,
      "n": 13,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "source=commons",
      "target": "self-gemini",
      "value": 0.6666666666666666,
      "lo": 0.5097739439934599,
      "hi": 0.7936704381040163,
      "n": 39,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "source=commons",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.08966685370757144,
      "n": 39,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "source=commons",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.08966685370757144,
      "n": 39,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "source=edited:gemini-nano-banana-2.1",
      "target": "self-gemini",
      "value": 0.6666666666666666,
      "lo": 0.39062208887279953,
      "hi": 0.8618799089087867,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "source=edited:gemini-nano-banana-2.1",
      "target": "self-gemini",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "source=edited:gemini-nano-banana-2.1",
      "target": "self-gemini",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "source=edited:gpt-image-2.5-sunburst",
      "target": "self-gemini",
      "value": 0.6666666666666666,
      "lo": 0.39062208887279953,
      "hi": 0.8618799089087867,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "source=edited:gpt-image-2.5-sunburst",
      "target": "self-gemini",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "source=edited:gpt-image-2.5-sunburst",
      "target": "self-gemini",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "source=edited:sharp-patch",
      "target": "self-gemini",
      "value": 5308.5,
      "lo": 5308.5,
      "hi": 5308.5,
      "n": 10,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "source=edited:sharp-patch",
      "target": "self-gemini",
      "value": 5896.299999999999,
      "lo": 5896.299999999999,
      "hi": 5896.299999999999,
      "n": 10,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "source=edited:sharp-patch",
      "target": "self-gemini",
      "value": 6006.46,
      "lo": 6006.46,
      "hi": 6006.46,
      "n": 10,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "source=edited:sharp-patch",
      "target": "self-gemini",
      "value": 0.01633333333333333,
      "lo": 0.01633333333333333,
      "hi": 0.01633333333333333,
      "n": 10,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "source=edited:sharp-patch",
      "target": "self-gemini",
      "value": 0.004750200000000001,
      "lo": 0.004750200000000001,
      "hi": 0.004750200000000001,
      "n": 10,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "source=edited:sharp-patch",
      "target": "self-gemini",
      "value": 0.6666666666666666,
      "lo": 0.4878005164454384,
      "hi": 0.8076950191632386,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "source=edited:sharp-patch",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "source=edited:sharp-patch",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "source=generated:flux-3-image",
      "target": "self-gemini",
      "value": 0.6666666666666666,
      "lo": 0.43749467295945105,
      "hi": 0.837212252491663,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "source=generated:flux-3-image",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.17587922364665765,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "source=generated:flux-3-image",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.17587922364665765,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "source=generated:gemini-nano-banana-2.1",
      "target": "self-gemini",
      "value": 0.6666666666666666,
      "lo": 0.4537345197425524,
      "hi": 0.8280524739491653,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "source=generated:gemini-nano-banana-2.1",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.154639018924847,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "source=generated:gemini-nano-banana-2.1",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.154639018924847,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "source=generated:gpt-image-2.5-flare",
      "target": "self-gemini",
      "value": 0.6666666666666666,
      "lo": 0.4537345197425524,
      "hi": 0.8280524739491653,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "source=generated:gpt-image-2.5-flare",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.154639018924847,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "source=generated:gpt-image-2.5-flare",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.154639018924847,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "source=sroie",
      "target": "self-gemini",
      "value": 4738,
      "lo": 4738,
      "hi": 4738,
      "n": 97,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "source=sroie",
      "target": "self-gemini",
      "value": 6229.399999999999,
      "lo": 6229.399999999999,
      "hi": 6229.399999999999,
      "n": 97,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "source=sroie",
      "target": "self-gemini",
      "value": 8481.199999999988,
      "lo": 8481.199999999988,
      "hi": 8481.199999999988,
      "n": 97,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "source=sroie",
      "target": "self-gemini",
      "value": 0.01633333333333333,
      "lo": 0.01633333333333333,
      "hi": 0.01633333333333333,
      "n": 97,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "source=sroie",
      "target": "self-gemini",
      "value": 0.004360817010309278,
      "lo": 0.004360817010309278,
      "hi": 0.004360817010309278,
      "n": 97,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "source=sroie",
      "target": "self-gemini",
      "value": 0.6700680272108843,
      "lo": 0.6144312470841656,
      "hi": 0.7213178470679745,
      "n": 294,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "source=sroie",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.012897663192708011,
      "n": 294,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "source=sroie",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.012897663192708011,
      "n": 294,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "transform=crop",
      "target": "self-gemini",
      "value": 0.7333333333333333,
      "lo": 0.4804956594401943,
      "hi": 0.8910254667430761,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "transform=crop",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "transform=crop",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "transform=jpeg",
      "target": "self-gemini",
      "value": 0.6666666666666666,
      "lo": 0.41713547738519774,
      "hi": 0.8482367556028526,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "transform=jpeg",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "transform=jpeg",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "transform=rephoto",
      "target": "self-gemini",
      "value": 0.6666666666666666,
      "lo": 0.41713547738519774,
      "hi": 0.8482367556028526,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "transform=rephoto",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "transform=rephoto",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "transform=resize",
      "target": "self-gemini",
      "value": 0.6666666666666666,
      "lo": 0.41713547738519774,
      "hi": 0.8482367556028526,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "transform=resize",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "transform=resize",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "transform=rotate",
      "target": "self-gemini",
      "value": 0.6666666666666666,
      "lo": 0.41713547738519774,
      "hi": 0.8482367556028526,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "transform=rotate",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "transform=rotate",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "transform=screenshot",
      "target": "self-gemini",
      "value": 0.6666666666666666,
      "lo": 0.41713547738519774,
      "hi": 0.8482367556028526,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "transform=screenshot",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "transform=screenshot",
      "target": "self-gemini",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "all",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9859649122807016,
      "lo": 0.9371692132513808,
      "hi": 0.9969867010277976,
      "n": 95,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "all",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9791666666666666,
      "lo": 0.9327980983028871,
      "hi": 0.9937556107579583,
      "n": 112,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "all",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.9581729333351078,
      "hi": 0.9999999999999999,
      "n": 88,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.subtotal",
      "stratum": "all",
      "target": "diy:claude-haiku-5.5",
      "value": 0.875,
      "lo": 0.639771727342413,
      "hi": 0.9650225122567595,
      "n": 16,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.tax",
      "stratum": "all",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.8388748419471806,
      "hi": 1,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "all",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9944903581267218,
      "lo": 0.9592714842947874,
      "hi": 0.999277605062941,
      "n": 121,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.receipt_number",
      "stratum": "all",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.7411670330319684,
      "hi": 1,
      "n": 11,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "all",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9504132231404959,
      "lo": 0.8960290618707025,
      "hi": 0.9770783259286766,
      "n": 121,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "line_items_f1",
      "stratum": "all",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9612403100775194,
      "lo": 0.9047332185886403,
      "hi": 1,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "all",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.9692293020525218,
      "hi": 1,
      "n": 121,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "arithmetic=consistent",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9822222222222221,
      "lo": 0.921271845462039,
      "hi": 0.9961811597878769,
      "n": 75,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "arithmetic=consistent",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9746376811594202,
      "lo": 0.9188345746837493,
      "hi": 0.9923925122987399,
      "n": 92,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "arithmetic=consistent",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.9500079920442688,
      "hi": 0.9999999999999998,
      "n": 73,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.subtotal",
      "stratum": "arithmetic=consistent",
      "target": "diy:claude-haiku-5.5",
      "value": 0.8181818181818182,
      "lo": 0.523019438039156,
      "hi": 0.9486323102539149,
      "n": 11,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.tax",
      "stratum": "arithmetic=consistent",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.7719046276458016,
      "hi": 0.9999999999999999,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "arithmetic=consistent",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.9630065017930143,
      "hi": 1,
      "n": 100,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "arithmetic=consistent",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9466666666666665,
      "lo": 0.8838741744259885,
      "hi": 0.9764116338424375,
      "n": 100,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "line_items_f1",
      "stratum": "arithmetic=consistent",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9880952380952381,
      "lo": 0.9607843137254902,
      "hi": 1,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "arithmetic=consistent",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.9630065017930143,
      "hi": 1,
      "n": 100,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "arithmetic=inconsistent",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.8156818649911483,
      "hi": 1,
      "n": 17,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "arithmetic=inconsistent",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.8241207763533425,
      "hi": 1,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "arithmetic=inconsistent",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.7846891972623643,
      "hi": 1,
      "n": 14,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "arithmetic=inconsistent",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9629629629629631,
      "lo": 0.7679464317540228,
      "hi": 0.9951283611657391,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "arithmetic=inconsistent",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9629629629629631,
      "lo": 0.7679464317540228,
      "hi": 0.9951283611657391,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "arithmetic=inconsistent",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.8241207763533425,
      "hi": 1,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "base=sroie",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.8156818649911483,
      "hi": 1,
      "n": 17,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "base=sroie",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.8388748419471806,
      "hi": 1,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "base=sroie",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.8063923194655637,
      "hi": 1,
      "n": 16,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "base=sroie",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9666666666666666,
      "lo": 0.7873326750830549,
      "hi": 0.9956171774009801,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "base=sroie",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9666666666666666,
      "lo": 0.7873326750830549,
      "hi": 0.9956171774009801,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "base=sroie",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.8388748419471806,
      "hi": 1,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "capture=photo",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.7846891972623643,
      "hi": 1,
      "n": 14,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "capture=photo",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.8453609810751531,
      "hi": 1,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "capture=photo",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.7719046276458016,
      "hi": 0.9999999999999999,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.tax",
      "stratum": "capture=photo",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.7575059933447592,
      "hi": 1,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "capture=photo",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.8453609810751531,
      "hi": 1,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "capture=photo",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.8453609810751531,
      "hi": 1,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "line_items_f1",
      "stratum": "capture=photo",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9743589743589743,
      "lo": 0.9333333333333333,
      "hi": 1,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "capture=photo",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.8453609810751531,
      "hi": 1,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "capture=scan",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9835390946502056,
      "lo": 0.9268260532345567,
      "hi": 0.9964646871908851,
      "n": 81,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "capture=scan",
      "target": "diy:claude-haiku-5.5",
      "value": 0.974074074074074,
      "lo": 0.9171123700538011,
      "hi": 0.9922227405222446,
      "n": 90,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "capture=scan",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.950650220603614,
      "hi": 1,
      "n": 74,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "capture=scan",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9932659932659934,
      "lo": 0.950564884217766,
      "hi": 0.9991169638095968,
      "n": 99,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "capture=scan",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9393939393939394,
      "lo": 0.8740321262351005,
      "hi": 0.9719302025537786,
      "n": 99,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "capture=scan",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.9626467879321726,
      "hi": 1,
      "n": 99,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "class=ai_full",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.8318207759473837,
      "hi": 1,
      "n": 19,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "class=ai_full",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.8388748419471806,
      "hi": 1,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "class=ai_full",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.7575059933447592,
      "hi": 1,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.subtotal",
      "stratum": "class=ai_full",
      "target": "diy:claude-haiku-5.5",
      "value": 0.875,
      "lo": 0.639771727342413,
      "hi": 0.9650225122567595,
      "n": 16,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.tax",
      "stratum": "class=ai_full",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.8388748419471806,
      "hi": 1,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "class=ai_full",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.8388748419471806,
      "hi": 1,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.receipt_number",
      "stratum": "class=ai_full",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.7411670330319684,
      "hi": 1,
      "n": 11,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "class=ai_full",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9,
      "lo": 0.6989663547715128,
      "hi": 0.9721335187862319,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "line_items_f1",
      "stratum": "class=ai_full",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9612403100775194,
      "lo": 0.9047332185886403,
      "hi": 1,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "class=ai_full",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.8388748419471806,
      "hi": 1,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "class=ai_inpaint",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.7224672001371107,
      "hi": 0.9999999999999999,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "class=ai_inpaint",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.7224672001371107,
      "hi": 0.9999999999999999,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "class=ai_inpaint",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.7224672001371107,
      "hi": 0.9999999999999999,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "class=ai_inpaint",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9666666666666666,
      "lo": 0.6767863445664513,
      "hi": 0.9975163755615184,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "class=ai_inpaint",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9666666666666666,
      "lo": 0.6767863445664513,
      "hi": 0.9975163755615184,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "class=ai_inpaint",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.7224672001371107,
      "hi": 0.9999999999999999,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "class=genuine",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9727891156462585,
      "lo": 0.8826896443704224,
      "hi": 0.9941471116047544,
      "n": 49,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "class=genuine",
      "target": "diy:claude-haiku-5.5",
      "value": 0.967741935483871,
      "lo": 0.8897953040501456,
      "hi": 0.9911086285001833,
      "n": 62,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "class=genuine",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.9299533801014687,
      "hi": 1,
      "n": 51,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "class=genuine",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.9479769375897331,
      "hi": 1,
      "n": 70,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "class=genuine",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9571428571428572,
      "lo": 0.8814033125147246,
      "hi": 0.9853184589958887,
      "n": 70,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "class=genuine",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.9479769375897331,
      "hi": 1,
      "n": 70,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "class=hard_negative",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.7224672001371107,
      "hi": 0.9999999999999999,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "class=hard_negative",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9666666666666668,
      "lo": 0.6767863445664516,
      "hi": 0.9975163755615185,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "class=hard_negative",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.7411670330319684,
      "hi": 1,
      "n": 11,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "class=hard_negative",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9696969696969698,
      "lo": 0.6985052183558994,
      "hi": 0.9977426005529196,
      "n": 11,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "class=hard_negative",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.7411670330319684,
      "hi": 1,
      "n": 11,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "class=manual_edit",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.7224672001371107,
      "hi": 0.9999999999999999,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "class=manual_edit",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9666666666666666,
      "lo": 0.6767863445664513,
      "hi": 0.9975163755615184,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "class=manual_edit",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9666666666666666,
      "lo": 0.6767863445664513,
      "hi": 0.9975163755615184,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "class=manual_edit",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.7224672001371107,
      "hi": 0.9999999999999999,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "country=MY",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9826839826839826,
      "lo": 0.9232146094426825,
      "hi": 0.9962805903928893,
      "n": 77,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "country=MY",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9722222222222221,
      "lo": 0.9114775589582506,
      "hi": 0.9916646975390148,
      "n": 84,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "country=MY",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.9486720478031097,
      "hi": 1,
      "n": 71,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "country=MY",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9928315412186381,
      "lo": 0.9475043370112963,
      "hi": 0.9990599525291086,
      "n": 93,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "country=MY",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9354838709677419,
      "lo": 0.8663200583761685,
      "hi": 0.9700985683813523,
      "n": 93,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "country=MY",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.960332497388265,
      "hi": 1,
      "n": 93,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "currency=MYR",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9826839826839826,
      "lo": 0.9232146094426825,
      "hi": 0.9962805903928893,
      "n": 77,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "currency=MYR",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9722222222222221,
      "lo": 0.9114775589582506,
      "hi": 0.9916646975390148,
      "n": 84,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "currency=MYR",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.9486720478031097,
      "hi": 1,
      "n": 71,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "currency=MYR",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9928315412186381,
      "lo": 0.9475043370112963,
      "hi": 0.9990599525291086,
      "n": 93,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "currency=MYR",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9354838709677419,
      "lo": 0.8663200583761685,
      "hi": 0.9700985683813523,
      "n": 93,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "currency=MYR",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.960332497388265,
      "hi": 1,
      "n": 93,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "dateEra=old",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9837398373983739,
      "lo": 0.9276764572931817,
      "hi": 0.9965078986489164,
      "n": 82,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "dateEra=old",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9737827715355805,
      "lo": 0.9162235735235813,
      "hi": 0.9921349809903817,
      "n": 89,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "dateEra=old",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.950650220603614,
      "hi": 1,
      "n": 74,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.tax",
      "stratum": "dateEra=old",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.7224672001371107,
      "hi": 0.9999999999999999,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "dateEra=old",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9931972789115647,
      "lo": 0.9500798177936901,
      "hi": 0.9991079470522714,
      "n": 98,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "dateEra=old",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9387755102040817,
      "lo": 0.8728091282684823,
      "hi": 0.9716406763186144,
      "n": 98,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "line_items_f1",
      "stratum": "dateEra=old",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9365079365079365,
      "lo": 0.8333333333333334,
      "hi": 1,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "dateEra=old",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.9622800098783192,
      "hi": 0.9999999999999999,
      "n": 98,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "dateEra=recent",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.7719046276458016,
      "hi": 0.9999999999999999,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "dateEra=recent",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.8568833815495731,
      "hi": 1,
      "n": 23,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "dateEra=recent",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.7846891972623643,
      "hi": 1,
      "n": 14,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.tax",
      "stratum": "dateEra=recent",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.7224672001371107,
      "hi": 0.9999999999999999,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "dateEra=recent",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.8568833815495731,
      "hi": 1,
      "n": 23,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "dateEra=recent",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.8568833815495731,
      "hi": 1,
      "n": 23,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "line_items_f1",
      "stratum": "dateEra=recent",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9848484848484849,
      "lo": 0.95,
      "hi": 1,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "dateEra=recent",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.8568833815495731,
      "hi": 1,
      "n": 23,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "docType=receipt",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9859649122807016,
      "lo": 0.9371692132513808,
      "hi": 0.9969867010277976,
      "n": 95,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "docType=receipt",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9791666666666666,
      "lo": 0.9327980983028871,
      "hi": 0.9937556107579583,
      "n": 112,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "docType=receipt",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.9581729333351078,
      "hi": 0.9999999999999999,
      "n": 88,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.subtotal",
      "stratum": "docType=receipt",
      "target": "diy:claude-haiku-5.5",
      "value": 0.875,
      "lo": 0.639771727342413,
      "hi": 0.9650225122567595,
      "n": 16,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.tax",
      "stratum": "docType=receipt",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.8388748419471806,
      "hi": 1,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "docType=receipt",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9944903581267218,
      "lo": 0.9592714842947874,
      "hi": 0.999277605062941,
      "n": 121,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.receipt_number",
      "stratum": "docType=receipt",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.7411670330319684,
      "hi": 1,
      "n": 11,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "docType=receipt",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9504132231404959,
      "lo": 0.8960290618707025,
      "hi": 0.9770783259286766,
      "n": 121,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "line_items_f1",
      "stratum": "docType=receipt",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9612403100775194,
      "lo": 0.9024390243902439,
      "hi": 1,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "docType=receipt",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.9692293020525218,
      "hi": 1,
      "n": 121,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "lang=en",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9839357429718875,
      "lo": 0.9285073245434474,
      "hi": 0.9965500665259674,
      "n": 83,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "lang=en",
      "target": "diy:claude-haiku-5.5",
      "value": 0.974074074074074,
      "lo": 0.9171123700538011,
      "hi": 0.9922227405222446,
      "n": 90,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "lang=en",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.950650220603614,
      "hi": 1,
      "n": 74,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.tax",
      "stratum": "lang=en",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.7224672001371107,
      "hi": 0.9999999999999999,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "lang=en",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9932659932659934,
      "lo": 0.950564884217766,
      "hi": 0.9991169638095968,
      "n": 99,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "lang=en",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9393939393939394,
      "lo": 0.8740321262351005,
      "hi": 0.9719302025537786,
      "n": 99,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "line_items_f1",
      "stratum": "lang=en",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9349593495934959,
      "lo": 0.8290598290598291,
      "hi": 1,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "lang=en",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.9626467879321726,
      "hi": 1,
      "n": 99,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "provenance=intact",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.8156818649911483,
      "hi": 1,
      "n": 17,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "provenance=intact",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.8388748419471806,
      "hi": 1,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "provenance=intact",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.7575059933447592,
      "hi": 1,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.tax",
      "stratum": "provenance=intact",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.7224672001371107,
      "hi": 0.9999999999999999,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "provenance=intact",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9833333333333332,
      "lo": 0.812152827904405,
      "hi": 0.9987595193112027,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "provenance=intact",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9333333333333332,
      "lo": 0.7414447301258664,
      "hi": 0.9855801328950236,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "line_items_f1",
      "stratum": "provenance=intact",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9534883720930233,
      "lo": 0.8536585365853658,
      "hi": 1,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "provenance=intact",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.8388748419471806,
      "hi": 1,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "provenance=none",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9774011299435028,
      "lo": 0.9012935311709601,
      "hi": 0.9951422675915979,
      "n": 59,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "provenance=none",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9675925925925924,
      "lo": 0.8975488579372277,
      "hi": 0.9902681010234919,
      "n": 72,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "provenance=none",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.93982814785791,
      "hi": 0.9999999999999998,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "provenance=none",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.9547219145675847,
      "hi": 1,
      "n": 81,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "provenance=none",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9588477366255143,
      "lo": 0.8909737038011953,
      "hi": 0.9851702754110323,
      "n": 81,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "provenance=none",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.9547219145675847,
      "hi": 1,
      "n": 81,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "provenance=stripped",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.8318207759473837,
      "hi": 1,
      "n": 19,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "provenance=stripped",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.8388748419471806,
      "hi": 1,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "provenance=stripped",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.8063923194655637,
      "hi": 1,
      "n": 16,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.tax",
      "stratum": "provenance=stripped",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.7224672001371107,
      "hi": 0.9999999999999999,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "provenance=stripped",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9833333333333332,
      "lo": 0.812152827904405,
      "hi": 0.9987595193112027,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "provenance=stripped",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9333333333333332,
      "lo": 0.7414447301258664,
      "hi": 0.9855801328950236,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "line_items_f1",
      "stratum": "provenance=stripped",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9689922480620154,
      "lo": 0.9206349206349206,
      "hi": 1,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "provenance=stripped",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.8388748419471806,
      "hi": 1,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "source=commons",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.7719046276458016,
      "hi": 0.9999999999999999,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "source=commons",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.7719046276458016,
      "hi": 0.9999999999999999,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "source=commons",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.7719046276458016,
      "hi": 0.9999999999999999,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "source=commons",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.7719046276458016,
      "hi": 0.9999999999999999,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "source=edited:sharp-patch",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.7224672001371107,
      "hi": 0.9999999999999999,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "source=edited:sharp-patch",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9666666666666666,
      "lo": 0.6767863445664513,
      "hi": 0.9975163755615184,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "source=edited:sharp-patch",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9666666666666666,
      "lo": 0.6767863445664513,
      "hi": 0.9975163755615184,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "source=edited:sharp-patch",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.7224672001371107,
      "hi": 0.9999999999999999,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "source=sroie",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9757575757575757,
      "lo": 0.8946078006966166,
      "hi": 0.9947877782404587,
      "n": 55,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "source=sroie",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9604519774011299,
      "lo": 0.8764998377475939,
      "hi": 0.9881098379819737,
      "n": 59,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "source=sroie",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.9312077638761376,
      "hi": 1,
      "n": 52,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "source=sroie",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.9465286634799296,
      "hi": 1,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "source=sroie",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9509803921568626,
      "lo": 0.8714206986988879,
      "hi": 0.9823110369888912,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "source=sroie",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.9465286634799296,
      "hi": 1,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "all",
      "target": "diy:claude-haiku-5.5",
      "value": 0.7956790123456791,
      "lo": 0.6958256172839506,
      "hi": 0.8862770061728394,
      "n": 121,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "all",
      "target": "diy:claude-haiku-5.5",
      "value": 0.475,
      "lo": 0.325,
      "hi": 0.65,
      "n": 40,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "all",
      "target": "diy:claude-haiku-5.5",
      "value": 0.575,
      "lo": 0.4,
      "hi": 0.725,
      "n": 40,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "all",
      "target": "diy:claude-haiku-5.5",
      "value": 0.625,
      "lo": 0.45,
      "hi": 0.75,
      "n": 40,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "all",
      "target": "diy:claude-haiku-5.5",
      "value": 0.475,
      "lo": 0.329354701126789,
      "hi": 0.6250263789330753,
      "n": 40,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_5",
      "stratum": "all",
      "target": "diy:claude-haiku-5.5",
      "value": 0.016460905349794237,
      "lo": 0.0035353128091148087,
      "hi": 0.07317394676544312,
      "n": 81,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "all",
      "target": "diy:claude-haiku-5.5",
      "value": 0.3083333333333333,
      "lo": 0.18740105428149992,
      "hi": 0.4628538928441266,
      "n": 40,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_7",
      "stratum": "all",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.045278085432415204,
      "n": 81,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "base=sroie",
      "target": "diy:claude-haiku-5.5",
      "value": 0.4666666666666666,
      "lo": 0.27171000474216656,
      "hi": 0.672365005794688,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "base=sroie",
      "target": "diy:claude-haiku-5.5",
      "value": 0.2833333333333333,
      "lo": 0.13402625504293345,
      "hi": 0.5024613134466217,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "capture=photo",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9351851851851852,
      "lo": 0.8148148148148148,
      "hi": 1,
      "n": 21,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "capture=photo",
      "target": "diy:claude-haiku-5.5",
      "value": 0.75,
      "lo": 0.5,
      "hi": 1,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "capture=photo",
      "target": "diy:claude-haiku-5.5",
      "value": 0.75,
      "lo": 0.5,
      "hi": 1,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "capture=photo",
      "target": "diy:claude-haiku-5.5",
      "value": 0.75,
      "lo": 0.5,
      "hi": 1,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "capture=photo",
      "target": "diy:claude-haiku-5.5",
      "value": 0.5555555555555555,
      "lo": 0.29701927644834225,
      "hi": 0.7871480561455196,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "capture=photo",
      "target": "diy:claude-haiku-5.5",
      "value": 0.3333333333333333,
      "lo": 0.13812009109121315,
      "hi": 0.6093779111272004,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "capture=scan",
      "target": "diy:claude-haiku-5.5",
      "value": 0.7718812877263581,
      "lo": 0.6483777665995976,
      "hi": 0.8833060865191147,
      "n": 99,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "capture=scan",
      "target": "diy:claude-haiku-5.5",
      "value": 0.42857142857142855,
      "lo": 0.25,
      "hi": 0.6428571428571429,
      "n": 28,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "capture=scan",
      "target": "diy:claude-haiku-5.5",
      "value": 0.5357142857142857,
      "lo": 0.2857142857142857,
      "hi": 0.7142857142857143,
      "n": 28,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "capture=scan",
      "target": "diy:claude-haiku-5.5",
      "value": 0.5714285714285714,
      "lo": 0.39285714285714285,
      "hi": 0.75,
      "n": 28,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "capture=scan",
      "target": "diy:claude-haiku-5.5",
      "value": 0.4404761904761905,
      "lo": 0.2750741835460622,
      "hi": 0.620240495592729,
      "n": 28,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_5",
      "stratum": "capture=scan",
      "target": "diy:claude-haiku-5.5",
      "value": 0.018779342723004695,
      "lo": 0.00403455081372734,
      "hi": 0.08292427641802158,
      "n": 71,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "capture=scan",
      "target": "diy:claude-haiku-5.5",
      "value": 0.2976190476190476,
      "lo": 0.16136246713704508,
      "hi": 0.4827074419348448,
      "n": 28,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_7",
      "stratum": "capture=scan",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.05132795219689034,
      "n": 71,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "country=MY",
      "target": "diy:claude-haiku-5.5",
      "value": 0.7685294117647059,
      "lo": 0.6320588235294118,
      "hi": 0.8867794117647058,
      "n": 93,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "country=MY",
      "target": "diy:claude-haiku-5.5",
      "value": 0.44,
      "lo": 0.24,
      "hi": 0.64,
      "n": 25,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "country=MY",
      "target": "diy:claude-haiku-5.5",
      "value": 0.56,
      "lo": 0.28,
      "hi": 0.76,
      "n": 25,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "country=MY",
      "target": "diy:claude-haiku-5.5",
      "value": 0.56,
      "lo": 0.36,
      "hi": 0.8,
      "n": 25,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "country=MY",
      "target": "diy:claude-haiku-5.5",
      "value": 0.45333333333333337,
      "lo": 0.2777616405486877,
      "hi": 0.6413363028722903,
      "n": 25,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_5",
      "stratum": "country=MY",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0196078431372549,
      "lo": 0.004213033667745614,
      "hi": 0.0863770739691849,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "country=MY",
      "target": "diy:claude-haiku-5.5",
      "value": 0.3066666666666667,
      "lo": 0.16217404611965835,
      "hi": 0.502660290910107,
      "n": 25,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_7",
      "stratum": "country=MY",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.053471336520070537,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "currency=MYR",
      "target": "diy:claude-haiku-5.5",
      "value": 0.7685294117647059,
      "lo": 0.6388014705882353,
      "hi": 0.887080882352941,
      "n": 93,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "currency=MYR",
      "target": "diy:claude-haiku-5.5",
      "value": 0.44,
      "lo": 0.24,
      "hi": 0.68,
      "n": 25,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "currency=MYR",
      "target": "diy:claude-haiku-5.5",
      "value": 0.56,
      "lo": 0.28,
      "hi": 0.76,
      "n": 25,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "currency=MYR",
      "target": "diy:claude-haiku-5.5",
      "value": 0.56,
      "lo": 0.4,
      "hi": 0.8,
      "n": 25,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "currency=MYR",
      "target": "diy:claude-haiku-5.5",
      "value": 0.45333333333333337,
      "lo": 0.2777616405486877,
      "hi": 0.6413363028722903,
      "n": 25,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_5",
      "stratum": "currency=MYR",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0196078431372549,
      "lo": 0.004213033667745614,
      "hi": 0.0863770739691849,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "currency=MYR",
      "target": "diy:claude-haiku-5.5",
      "value": 0.3066666666666667,
      "lo": 0.16217404611965835,
      "hi": 0.502660290910107,
      "n": 25,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_7",
      "stratum": "currency=MYR",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.053471336520070537,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "dateEra=old",
      "target": "diy:claude-haiku-5.5",
      "value": 0.7568627450980392,
      "lo": 0.6316115196078431,
      "hi": 0.8715747549019608,
      "n": 98,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "dateEra=old",
      "target": "diy:claude-haiku-5.5",
      "value": 0.4,
      "lo": 0.23333333333333334,
      "hi": 0.6,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "dateEra=old",
      "target": "diy:claude-haiku-5.5",
      "value": 0.5333333333333333,
      "lo": 0.3,
      "hi": 0.7,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "dateEra=old",
      "target": "diy:claude-haiku-5.5",
      "value": 0.5333333333333333,
      "lo": 0.36666666666666664,
      "hi": 0.7333333333333333,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "dateEra=old",
      "target": "diy:claude-haiku-5.5",
      "value": 0.41111111111111115,
      "lo": 0.25511909619678397,
      "hi": 0.5872832848119216,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_5",
      "stratum": "dateEra=old",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0196078431372549,
      "lo": 0.004213033667745614,
      "hi": 0.0863770739691849,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "dateEra=old",
      "target": "diy:claude-haiku-5.5",
      "value": 0.26666666666666666,
      "lo": 0.14182663319596317,
      "hi": 0.4444796169518888,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_7",
      "stratum": "dateEra=old",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.053471336520070537,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "dateEra=recent",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9192307692307692,
      "lo": 0.7769230769230769,
      "hi": 1,
      "n": 23,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "dateEra=recent",
      "target": "diy:claude-haiku-5.5",
      "value": 0.7,
      "lo": 0.4,
      "hi": 1,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "dateEra=recent",
      "target": "diy:claude-haiku-5.5",
      "value": 0.7,
      "lo": 0.4,
      "hi": 1,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "dateEra=recent",
      "target": "diy:claude-haiku-5.5",
      "value": 0.8,
      "lo": 0.5,
      "hi": 1,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "dateEra=recent",
      "target": "diy:claude-haiku-5.5",
      "value": 0.6666666666666666,
      "lo": 0.36779781217588203,
      "hi": 0.8730245878698215,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_5",
      "stratum": "dateEra=recent",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.22809537235419838,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "dateEra=recent",
      "target": "diy:claude-haiku-5.5",
      "value": 0.4333333333333334,
      "lo": 0.190125658185213,
      "hi": 0.7135453817965056,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_7",
      "stratum": "dateEra=recent",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.22809537235419838,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "docType=receipt",
      "target": "diy:claude-haiku-5.5",
      "value": 0.7956790123456791,
      "lo": 0.6976195987654322,
      "hi": 0.8821141975308641,
      "n": 121,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "docType=receipt",
      "target": "diy:claude-haiku-5.5",
      "value": 0.475,
      "lo": 0.325,
      "hi": 0.625,
      "n": 40,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "docType=receipt",
      "target": "diy:claude-haiku-5.5",
      "value": 0.575,
      "lo": 0.4,
      "hi": 0.725,
      "n": 40,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "docType=receipt",
      "target": "diy:claude-haiku-5.5",
      "value": 0.625,
      "lo": 0.45,
      "hi": 0.7506249999999965,
      "n": 40,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "docType=receipt",
      "target": "diy:claude-haiku-5.5",
      "value": 0.475,
      "lo": 0.329354701126789,
      "hi": 0.6250263789330753,
      "n": 40,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_5",
      "stratum": "docType=receipt",
      "target": "diy:claude-haiku-5.5",
      "value": 0.016460905349794237,
      "lo": 0.0035353128091148087,
      "hi": 0.07317394676544312,
      "n": 81,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "docType=receipt",
      "target": "diy:claude-haiku-5.5",
      "value": 0.3083333333333333,
      "lo": 0.18740105428149992,
      "hi": 0.4628538928441266,
      "n": 40,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_7",
      "stratum": "docType=receipt",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.045278085432415204,
      "n": 81,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "lang=en",
      "target": "diy:claude-haiku-5.5",
      "value": 0.7478260869565218,
      "lo": 0.6330857487922705,
      "hi": 0.8529045893719807,
      "n": 99,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "lang=en",
      "target": "diy:claude-haiku-5.5",
      "value": 0.4,
      "lo": 0.23333333333333334,
      "hi": 0.5674999999999955,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "lang=en",
      "target": "diy:claude-haiku-5.5",
      "value": 0.5,
      "lo": 0.26666666666666666,
      "hi": 0.6666666666666666,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "lang=en",
      "target": "diy:claude-haiku-5.5",
      "value": 0.5,
      "lo": 0.3333333333333333,
      "hi": 0.7,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "lang=en",
      "target": "diy:claude-haiku-5.5",
      "value": 0.41111111111111115,
      "lo": 0.25511909619678397,
      "hi": 0.5872832848119216,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_5",
      "stratum": "lang=en",
      "target": "diy:claude-haiku-5.5",
      "value": 0.01932367149758454,
      "lo": 0.004151810238993414,
      "hi": 0.08519463603937033,
      "n": 69,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "lang=en",
      "target": "diy:claude-haiku-5.5",
      "value": 0.2888888888888889,
      "lo": 0.15827739295395987,
      "hi": 0.4674282619417158,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_7",
      "stratum": "lang=en",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 3.469446951953614e-18,
      "hi": 0.05273725818905173,
      "n": 69,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_5",
      "stratum": "source=commons",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.22809537235419838,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_7",
      "stratum": "source=commons",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.22809537235419838,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "source=edited:sharp-patch",
      "target": "diy:claude-haiku-5.5",
      "value": 0.4666666666666666,
      "lo": 0.21293408575713324,
      "hi": 0.7389014342337259,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "source=edited:sharp-patch",
      "target": "diy:claude-haiku-5.5",
      "value": 0.3333333333333333,
      "lo": 0.12697541213017843,
      "hi": 0.632202187824118,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_5",
      "stratum": "source=sroie",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0196078431372549,
      "lo": 0.004213033667745614,
      "hi": 0.0863770739691849,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_7",
      "stratum": "source=sroie",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.053471336520070537,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "provenance=intact",
      "target": "diy:claude-haiku-5.5",
      "value": 0.4,
      "lo": 0.21880653237281705,
      "hi": 0.6134184992377469,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "provenance=intact",
      "target": "diy:claude-haiku-5.5",
      "value": 0.3166666666666667,
      "lo": 0.15716056329227931,
      "hi": 0.5352519946604213,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "provenance=stripped",
      "target": "diy:claude-haiku-5.5",
      "value": 0.55,
      "lo": 0.34208534245034233,
      "hi": 0.7418021417443759,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "provenance=stripped",
      "target": "diy:claude-haiku-5.5",
      "value": 0.3,
      "lo": 0.1454772448676043,
      "hi": 0.5189728183535235,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_5",
      "stratum": "provenance=none",
      "target": "diy:claude-haiku-5.5",
      "value": 0.016460905349794237,
      "lo": 0.0035353128091148087,
      "hi": 0.07317394676544312,
      "n": 81,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_7",
      "stratum": "provenance=none",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.045278085432415204,
      "n": 81,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "class=ai_full",
      "target": "diy:claude-haiku-5.5",
      "value": 0.799074074074074,
      "lo": 0.6545987654320988,
      "hi": 0.9092592592592592,
      "n": 101,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "class=ai_full",
      "target": "diy:claude-haiku-5.5",
      "value": 0.5,
      "lo": 0.3,
      "hi": 0.7,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "class=ai_full",
      "target": "diy:claude-haiku-5.5",
      "value": 0.55,
      "lo": 0.35,
      "hi": 0.75,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "class=ai_full",
      "target": "diy:claude-haiku-5.5",
      "value": 0.55,
      "lo": 0.34875000000000217,
      "hi": 0.75,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "class=ai_full",
      "target": "diy:claude-haiku-5.5",
      "value": 0.4833333333333333,
      "lo": 0.2854103181114389,
      "hi": 0.6866271871569883,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "class=ai_full",
      "target": "diy:claude-haiku-5.5",
      "value": 0.33333333333333337,
      "lo": 0.16906782991982028,
      "hi": 0.551307222764453,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "class=ai_inpaint",
      "target": "diy:claude-haiku-5.5",
      "value": 0.7672839506172839,
      "lo": 0.528395061728395,
      "hi": 0.965555555555555,
      "n": 91,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "class=ai_inpaint",
      "target": "diy:claude-haiku-5.5",
      "value": 0.4,
      "lo": 0.1,
      "hi": 0.8,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "class=ai_inpaint",
      "target": "diy:claude-haiku-5.5",
      "value": 0.7,
      "lo": 0.2,
      "hi": 0.9,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "class=ai_inpaint",
      "target": "diy:claude-haiku-5.5",
      "value": 0.7,
      "lo": 0.4,
      "hi": 1,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "class=ai_inpaint",
      "target": "diy:claude-haiku-5.5",
      "value": 0.4666666666666666,
      "lo": 0.21293408575713324,
      "hi": 0.7389014342337259,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "class=ai_inpaint",
      "target": "diy:claude-haiku-5.5",
      "value": 0.2333333333333333,
      "lo": 0.07255530696407597,
      "hi": 0.5421288529627982,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "class=manual_edit",
      "target": "diy:claude-haiku-5.5",
      "value": 0.817283950617284,
      "lo": 0.6364043209876543,
      "hi": 0.9629629629629629,
      "n": 91,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "class=manual_edit",
      "target": "diy:claude-haiku-5.5",
      "value": 0.5,
      "lo": 0.2,
      "hi": 0.8,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "class=manual_edit",
      "target": "diy:claude-haiku-5.5",
      "value": 0.5,
      "lo": 0.2,
      "hi": 0.9,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "class=manual_edit",
      "target": "diy:claude-haiku-5.5",
      "value": 0.7,
      "lo": 0.3,
      "hi": 0.9,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "class=manual_edit",
      "target": "diy:claude-haiku-5.5",
      "value": 0.4666666666666666,
      "lo": 0.21293408575713324,
      "hi": 0.7389014342337259,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "class=manual_edit",
      "target": "diy:claude-haiku-5.5",
      "value": 0.3333333333333333,
      "lo": 0.12697541213017843,
      "hi": 0.632202187824118,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "arithmetic=consistent",
      "target": "diy:claude-haiku-5.5",
      "value": 0.6318742985409652,
      "lo": 0.4887696408529742,
      "hi": 0.756502525252525,
      "n": 103,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "arithmetic=consistent",
      "target": "diy:claude-haiku-5.5",
      "value": 0.13636363636363635,
      "lo": 0,
      "hi": 0.3181818181818182,
      "n": 22,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "arithmetic=consistent",
      "target": "diy:claude-haiku-5.5",
      "value": 0.22727272727272727,
      "lo": 0.045454545454545456,
      "hi": 0.45454545454545453,
      "n": 22,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "arithmetic=consistent",
      "target": "diy:claude-haiku-5.5",
      "value": 0.3181818181818182,
      "lo": 0.09090909090909091,
      "hi": 0.5,
      "n": 22,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "arithmetic=consistent",
      "target": "diy:claude-haiku-5.5",
      "value": 0.15151515151515152,
      "lo": 0.05568961722175389,
      "hi": 0.3509486287175879,
      "n": 22,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "arithmetic=consistent",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0606060606060606,
      "lo": 0.013097648773984727,
      "hi": 0.23875057436692448,
      "n": 22,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "arithmetic=inconsistent",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9958847736625515,
      "lo": 0.9862825788751715,
      "hi": 1,
      "n": 99,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "arithmetic=inconsistent",
      "target": "diy:claude-haiku-5.5",
      "value": 0.8888888888888888,
      "lo": 0.7222222222222222,
      "hi": 1,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "arithmetic=inconsistent",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 0.8333333333333334,
      "hi": 1,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "arithmetic=inconsistent",
      "target": "diy:claude-haiku-5.5",
      "value": 1,
      "lo": 1,
      "hi": 1,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "arithmetic=inconsistent",
      "target": "diy:claude-haiku-5.5",
      "value": 0.8703703703703703,
      "lo": 0.650030075749184,
      "hi": 0.9604297585866252,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "arithmetic=inconsistent",
      "target": "diy:claude-haiku-5.5",
      "value": 0.611111111111111,
      "lo": 0.38619041602237897,
      "hi": 0.7969475342783636,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "decisions",
      "metric": "decision_consistency",
      "stratum": "all",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9090909090909091,
      "lo": 0.8445270273318659,
      "hi": 0.9484787652565609,
      "n": 121,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "decisions",
      "metric": "total_consistency",
      "stratum": "all",
      "target": "diy:claude-haiku-5.5",
      "value": 0.9834710743801653,
      "lo": 0.9417332575142658,
      "hi": 0.9954554064538752,
      "n": 121,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "decisions",
      "metric": "calibration_error",
      "stratum": "all",
      "target": "diy:claude-haiku-5.5",
      "value": 0.10884297520661158,
      "lo": 0.08622314049586782,
      "hi": 0.1516108815426997,
      "n": 121,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_5",
      "stratum": "all",
      "target": "diy:claude-haiku-5.5",
      "value": 802.0370370370371,
      "lo": 494.2592592592593,
      "hi": 1195.8055555555557,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_7",
      "stratum": "all",
      "target": "diy:claude-haiku-5.5",
      "value": 691.6666666666666,
      "lo": 550,
      "hi": 825.0000000000001,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_5",
      "stratum": "assumption=fakes_0_1pct",
      "target": "diy:claude-haiku-5.5",
      "value": 332.0555555555555,
      "lo": 53.33333333333334,
      "hi": 744.7430555555553,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_7",
      "stratum": "assumption=fakes_0_1pct",
      "target": "diy:claude-haiku-5.5",
      "value": 69.16666666666667,
      "lo": 56.66666666666667,
      "hi": 81.66666666666667,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_5",
      "stratum": "assumption=fakes_5pct",
      "target": "diy:claude-haiku-5.5",
      "value": 2890.843621399177,
      "lo": 2166.640946502058,
      "hi": 3682.5102880658437,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_7",
      "stratum": "assumption=fakes_5pct",
      "target": "diy:claude-haiku-5.5",
      "value": 3458.3333333333335,
      "lo": 2791.666666666667,
      "hi": 4083.333333333333,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_5",
      "stratum": "assumption=fake_32usd",
      "target": "diy:claude-haiku-5.5",
      "value": 445.037037037037,
      "lo": 170.66666666666666,
      "hi": 839.4648148148141,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_7",
      "stratum": "assumption=fake_32usd",
      "target": "diy:claude-haiku-5.5",
      "value": 221.33333333333331,
      "lo": 176.00000000000003,
      "hi": 261.3333333333333,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_5",
      "stratum": "assumption=fake_182usd",
      "target": "diy:claude-haiku-5.5",
      "value": 1232.537037037037,
      "lo": 857.7615740740742,
      "hi": 1663.3226851851846,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_7",
      "stratum": "assumption=fake_182usd",
      "target": "diy:claude-haiku-5.5",
      "value": 1258.8333333333333,
      "lo": 1031.3333333333333,
      "hi": 1486.3333333333333,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_5",
      "stratum": "assumption=review_8usd",
      "target": "diy:claude-haiku-5.5",
      "value": 655.3703703703704,
      "lo": 465.1851851851852,
      "hi": 868.3333333333333,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_7",
      "stratum": "assumption=review_8usd",
      "target": "diy:claude-haiku-5.5",
      "value": 691.6666666666666,
      "lo": 566.4583333333337,
      "hi": 808.3333333333334,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_5",
      "stratum": "assumption=review_73usd",
      "target": "diy:claude-haiku-5.5",
      "value": 1714.6296296296298,
      "lo": 533.3333333333334,
      "hi": 3285.208333333332,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_7",
      "stratum": "assumption=review_73usd",
      "target": "diy:claude-haiku-5.5",
      "value": 691.6666666666666,
      "lo": 550,
      "hi": 808.3333333333334,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "approve_all_cost_per_1000",
      "stratum": "all",
      "target": "diy:claude-haiku-5.5",
      "value": 1000,
      "lo": 1000,
      "hi": 1000,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "approve_all_cost_per_1000",
      "stratum": "assumption=fakes_0_1pct",
      "target": "diy:claude-haiku-5.5",
      "value": 100,
      "lo": 100,
      "hi": 100,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "approve_all_cost_per_1000",
      "stratum": "assumption=fakes_5pct",
      "target": "diy:claude-haiku-5.5",
      "value": 5000,
      "lo": 5000,
      "hi": 5000,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "approve_all_cost_per_1000",
      "stratum": "assumption=fake_32usd",
      "target": "diy:claude-haiku-5.5",
      "value": 320,
      "lo": 320,
      "hi": 320,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "approve_all_cost_per_1000",
      "stratum": "assumption=fake_182usd",
      "target": "diy:claude-haiku-5.5",
      "value": 1820,
      "lo": 1820,
      "hi": 1820,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "approve_all_cost_per_1000",
      "stratum": "assumption=review_8usd",
      "target": "diy:claude-haiku-5.5",
      "value": 1000,
      "lo": 1000,
      "hi": 1000,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "approve_all_cost_per_1000",
      "stratum": "assumption=review_73usd",
      "target": "diy:claude-haiku-5.5",
      "value": 1000,
      "lo": 1000,
      "hi": 1000,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "evidence",
      "metric": "edited_field_named",
      "stratum": "all",
      "target": "diy:claude-haiku-5.5",
      "value": 0.4166666666666667,
      "lo": 0.23174394158041697,
      "hi": 0.6284435847617197,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "all",
      "target": "diy:claude-haiku-5.5",
      "value": 8424,
      "lo": 8424,
      "hi": 8424,
      "n": 453,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "all",
      "target": "diy:claude-haiku-5.5",
      "value": 13234.599999999999,
      "lo": 13234.599999999999,
      "hi": 13234.599999999999,
      "n": 453,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "all",
      "target": "diy:claude-haiku-5.5",
      "value": 15886.080000000002,
      "lo": 15886.080000000002,
      "hi": 15886.080000000002,
      "n": 453,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "all",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0008693229580573947,
      "lo": 0.0008693229580573947,
      "hi": 0.0008693229580573947,
      "n": 453,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "all",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0008693229580573947,
      "lo": 0.0008693229580573947,
      "hi": 0.0008693229580573947,
      "n": 453,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "all",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.008408735123582253,
      "n": 453,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "all",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.008408735123582253,
      "n": 453,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "all",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.008408735123582253,
      "n": 453,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "arithmetic=consistent",
      "target": "diy:claude-haiku-5.5",
      "value": 8235,
      "lo": 8235,
      "hi": 8235,
      "n": 390,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "arithmetic=consistent",
      "target": "diy:claude-haiku-5.5",
      "value": 12757.199999999999,
      "lo": 12757.199999999999,
      "hi": 12757.199999999999,
      "n": 390,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "arithmetic=consistent",
      "target": "diy:claude-haiku-5.5",
      "value": 14419.130000000021,
      "lo": 14419.130000000021,
      "hi": 14419.130000000021,
      "n": 390,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "arithmetic=consistent",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0008349556410256403,
      "lo": 0.0008349556410256403,
      "hi": 0.0008349556410256403,
      "n": 390,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "arithmetic=consistent",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0008349556410256403,
      "lo": 0.0008349556410256403,
      "hi": 0.0008349556410256403,
      "n": 390,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "arithmetic=consistent",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.009753820311850517,
      "n": 390,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "arithmetic=consistent",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.009753820311850517,
      "n": 390,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "arithmetic=consistent",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.009753820311850517,
      "n": 390,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "arithmetic=inconsistent",
      "target": "diy:claude-haiku-5.5",
      "value": 9527,
      "lo": 9527,
      "hi": 9527,
      "n": 54,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "arithmetic=inconsistent",
      "target": "diy:claude-haiku-5.5",
      "value": 14781.349999999997,
      "lo": 14781.349999999997,
      "hi": 14781.349999999997,
      "n": 54,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "arithmetic=inconsistent",
      "target": "diy:claude-haiku-5.5",
      "value": 15909.189999999999,
      "lo": 15909.189999999999,
      "hi": 15909.189999999999,
      "n": 54,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "arithmetic=inconsistent",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0010817866666666667,
      "lo": 0.0010817866666666667,
      "hi": 0.0010817866666666667,
      "n": 54,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "arithmetic=inconsistent",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0010817866666666667,
      "lo": 0.0010817866666666667,
      "hi": 0.0010817866666666667,
      "n": 54,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "arithmetic=inconsistent",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.0664135880908964,
      "n": 54,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "arithmetic=inconsistent",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.0664135880908964,
      "n": 54,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "arithmetic=inconsistent",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.0664135880908964,
      "n": 54,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "arithmeticError=totals",
      "target": "diy:claude-haiku-5.5",
      "value": 10072,
      "lo": 10072,
      "hi": 10072,
      "n": 21,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "arithmeticError=totals",
      "target": "diy:claude-haiku-5.5",
      "value": 12021,
      "lo": 12021,
      "hi": 12021,
      "n": 21,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "arithmeticError=totals",
      "target": "diy:claude-haiku-5.5",
      "value": 14510.600000000002,
      "lo": 14510.600000000002,
      "hi": 14510.600000000002,
      "n": 21,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "arithmeticError=totals",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0010411276190476187,
      "lo": 0.0010411276190476187,
      "hi": 0.0010411276190476187,
      "n": 21,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "arithmeticError=totals",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0010411276190476187,
      "lo": 0.0010411276190476187,
      "hi": 0.0010411276190476187,
      "n": 21,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "arithmeticError=totals",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.154639018924847,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "arithmeticError=totals",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.154639018924847,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "arithmeticError=totals",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.154639018924847,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "base=sroie",
      "target": "diy:claude-haiku-5.5",
      "value": 9094,
      "lo": 9094,
      "hi": 9094,
      "n": 60,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "base=sroie",
      "target": "diy:claude-haiku-5.5",
      "value": 14562.55,
      "lo": 14562.55,
      "hi": 14562.55,
      "n": 60,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "base=sroie",
      "target": "diy:claude-haiku-5.5",
      "value": 15874.569999999998,
      "lo": 15874.569999999998,
      "hi": 15874.569999999998,
      "n": 60,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "base=sroie",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0010004554999999999,
      "lo": 0.0010004554999999999,
      "hi": 0.0010004554999999999,
      "n": 60,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "base=sroie",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0010004554999999999,
      "lo": 0.0010004554999999999,
      "hi": 0.0010004554999999999,
      "n": 60,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "base=sroie",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.06017185214208986,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "base=sroie",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.06017185214208986,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "base=sroie",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.06017185214208986,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "capture=photo",
      "target": "diy:claude-haiku-5.5",
      "value": 8719,
      "lo": 8719,
      "hi": 8719,
      "n": 75,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "capture=photo",
      "target": "diy:claude-haiku-5.5",
      "value": 12999.4,
      "lo": 12999.4,
      "hi": 12999.4,
      "n": 75,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "capture=photo",
      "target": "diy:claude-haiku-5.5",
      "value": 15316.820000000003,
      "lo": 15316.820000000003,
      "hi": 15316.820000000003,
      "n": 75,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "capture=photo",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0008811044,
      "lo": 0.0008811044,
      "hi": 0.0008811044,
      "n": 75,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "capture=photo",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0008811044,
      "lo": 0.0008811044,
      "hi": 0.0008811044,
      "n": 75,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "capture=photo",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 3.469446951953614e-18,
      "hi": 0.04872384248280587,
      "n": 75,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "capture=photo",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 3.469446951953614e-18,
      "hi": 0.04872384248280587,
      "n": 75,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "capture=photo",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 3.469446951953614e-18,
      "hi": 0.04872384248280587,
      "n": 75,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "capture=scan",
      "target": "diy:claude-haiku-5.5",
      "value": 8368.5,
      "lo": 8368.5,
      "hi": 8368.5,
      "n": 360,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "capture=scan",
      "target": "diy:claude-haiku-5.5",
      "value": 13198.2,
      "lo": 13198.2,
      "hi": 13198.2,
      "n": 360,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "capture=scan",
      "target": "diy:claude-haiku-5.5",
      "value": 16050.390000000007,
      "lo": 16050.390000000007,
      "hi": 16050.390000000007,
      "n": 360,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "capture=scan",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0008612274166666657,
      "lo": 0.0008612274166666657,
      "hi": 0.0008612274166666657,
      "n": 360,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "capture=scan",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0008612274166666657,
      "lo": 0.0008612274166666657,
      "hi": 0.0008612274166666657,
      "n": 360,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "capture=scan",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.010558056888693507,
      "n": 360,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "capture=scan",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.010558056888693507,
      "n": 360,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "capture=scan",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.010558056888693507,
      "n": 360,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "capture=screenshot",
      "target": "diy:claude-haiku-5.5",
      "value": 7528,
      "lo": 7528,
      "hi": 7528,
      "n": 15,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "capture=screenshot",
      "target": "diy:claude-haiku-5.5",
      "value": 11765.099999999997,
      "lo": 11765.099999999997,
      "hi": 11765.099999999997,
      "n": 15,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "capture=screenshot",
      "target": "diy:claude-haiku-5.5",
      "value": 13118.619999999999,
      "lo": 13118.619999999999,
      "hi": 13118.619999999999,
      "n": 15,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "capture=screenshot",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0008422473333333332,
      "lo": 0.0008422473333333332,
      "hi": 0.0008422473333333332,
      "n": 15,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "capture=screenshot",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0008422473333333332,
      "lo": 0.0008422473333333332,
      "hi": 0.0008422473333333332,
      "n": 15,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "capture=screenshot",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "capture=screenshot",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "capture=screenshot",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "class=ai_full",
      "target": "diy:claude-haiku-5.5",
      "value": 8935.5,
      "lo": 8935.5,
      "hi": 8935.5,
      "n": 60,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "class=ai_full",
      "target": "diy:claude-haiku-5.5",
      "value": 11882.8,
      "lo": 11882.8,
      "hi": 11882.8,
      "n": 60,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "class=ai_full",
      "target": "diy:claude-haiku-5.5",
      "value": 13296.91999999999,
      "lo": 13296.91999999999,
      "hi": 13296.91999999999,
      "n": 60,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "class=ai_full",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0009271501666666665,
      "lo": 0.0009271501666666665,
      "hi": 0.0009271501666666665,
      "n": 60,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "class=ai_full",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0009271501666666665,
      "lo": 0.0009271501666666665,
      "hi": 0.0009271501666666665,
      "n": 60,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "class=ai_full",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.06017185214208986,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "class=ai_full",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.06017185214208986,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "class=ai_full",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.06017185214208986,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "class=ai_inpaint",
      "target": "diy:claude-haiku-5.5",
      "value": 8200,
      "lo": 8200,
      "hi": 8200,
      "n": 30,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "class=ai_inpaint",
      "target": "diy:claude-haiku-5.5",
      "value": 11633.499999999998,
      "lo": 11633.499999999998,
      "hi": 11633.499999999998,
      "n": 30,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "class=ai_inpaint",
      "target": "diy:claude-haiku-5.5",
      "value": 13093.240000000002,
      "lo": 13093.240000000002,
      "hi": 13093.240000000002,
      "n": 30,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "class=ai_inpaint",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0008051406666666664,
      "lo": 0.0008051406666666664,
      "hi": 0.0008051406666666664,
      "n": 30,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "class=ai_inpaint",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0008051406666666664,
      "lo": 0.0008051406666666664,
      "hi": 0.0008051406666666664,
      "n": 30,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "class=ai_inpaint",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "class=ai_inpaint",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "class=ai_inpaint",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "class=duplicate",
      "target": "diy:claude-haiku-5.5",
      "value": 8393.5,
      "lo": 8393.5,
      "hi": 8393.5,
      "n": 90,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "class=duplicate",
      "target": "diy:claude-haiku-5.5",
      "value": 12156.049999999997,
      "lo": 12156.049999999997,
      "hi": 12156.049999999997,
      "n": 90,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "class=duplicate",
      "target": "diy:claude-haiku-5.5",
      "value": 13729.689999999999,
      "lo": 13729.689999999999,
      "hi": 13729.689999999999,
      "n": 90,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "class=duplicate",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0008468908888888887,
      "lo": 0.0008468908888888887,
      "hi": 0.0008468908888888887,
      "n": 90,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "class=duplicate",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0008468908888888887,
      "lo": 0.0008468908888888887,
      "hi": 0.0008468908888888887,
      "n": 90,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "class=duplicate",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.04093562556432678,
      "n": 90,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "class=duplicate",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.04093562556432678,
      "n": 90,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "class=duplicate",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.04093562556432678,
      "n": 90,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "class=genuine",
      "target": "diy:claude-haiku-5.5",
      "value": 8089.5,
      "lo": 8089.5,
      "hi": 8089.5,
      "n": 210,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "class=genuine",
      "target": "diy:claude-haiku-5.5",
      "value": 13004.35,
      "lo": 13004.35,
      "hi": 13004.35,
      "n": 210,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "class=genuine",
      "target": "diy:claude-haiku-5.5",
      "value": 16679.929999999997,
      "lo": 16679.929999999997,
      "hi": 16679.929999999997,
      "n": 210,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "class=genuine",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0008326404761904758,
      "lo": 0.0008326404761904758,
      "hi": 0.0008326404761904758,
      "n": 210,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "class=genuine",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0008326404761904758,
      "lo": 0.0008326404761904758,
      "hi": 0.0008326404761904758,
      "n": 210,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "class=genuine",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.017964050759283236,
      "n": 210,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "class=genuine",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.017964050759283236,
      "n": 210,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "class=genuine",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.017964050759283236,
      "n": 210,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "class=hard_negative",
      "target": "diy:claude-haiku-5.5",
      "value": 7815,
      "lo": 7815,
      "hi": 7815,
      "n": 33,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "class=hard_negative",
      "target": "diy:claude-haiku-5.5",
      "value": 11744.999999999998,
      "lo": 11744.999999999998,
      "hi": 11744.999999999998,
      "n": 33,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "class=hard_negative",
      "target": "diy:claude-haiku-5.5",
      "value": 13389.8,
      "lo": 13389.8,
      "hi": 13389.8,
      "n": 33,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "class=hard_negative",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0008203721212121211,
      "lo": 0.0008203721212121211,
      "hi": 0.0008203721212121211,
      "n": 33,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "class=hard_negative",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0008203721212121211,
      "lo": 0.0008203721212121211,
      "hi": 0.0008203721212121211,
      "n": 33,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "class=hard_negative",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.10426999754245206,
      "n": 33,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "class=hard_negative",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.10426999754245206,
      "n": 33,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "class=hard_negative",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.10426999754245206,
      "n": 33,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "class=manual_edit",
      "target": "diy:claude-haiku-5.5",
      "value": 11458.5,
      "lo": 11458.5,
      "hi": 11458.5,
      "n": 30,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "class=manual_edit",
      "target": "diy:claude-haiku-5.5",
      "value": 15167.299999999997,
      "lo": 15167.299999999997,
      "hi": 15167.299999999997,
      "n": 30,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "class=manual_edit",
      "target": "diy:claude-haiku-5.5",
      "value": 16047.67,
      "lo": 16047.67,
      "hi": 16047.67,
      "n": 30,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "class=manual_edit",
      "target": "diy:claude-haiku-5.5",
      "value": 0.001195770333333333,
      "lo": 0.001195770333333333,
      "hi": 0.001195770333333333,
      "n": 30,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "class=manual_edit",
      "target": "diy:claude-haiku-5.5",
      "value": 0.001195770333333333,
      "lo": 0.001195770333333333,
      "hi": 0.001195770333333333,
      "n": 30,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "class=manual_edit",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "class=manual_edit",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "class=manual_edit",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "country=JP",
      "target": "diy:claude-haiku-5.5",
      "value": 3595,
      "lo": 3595,
      "hi": 3595,
      "n": 12,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "country=JP",
      "target": "diy:claude-haiku-5.5",
      "value": 8550.5,
      "lo": 8550.5,
      "hi": 8550.5,
      "n": 12,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "country=JP",
      "target": "diy:claude-haiku-5.5",
      "value": 8959.7,
      "lo": 8959.7,
      "hi": 8959.7,
      "n": 12,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "country=JP",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0004900783333333333,
      "lo": 0.0004900783333333333,
      "hi": 0.0004900783333333333,
      "n": 12,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "country=JP",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0004900783333333333,
      "lo": 0.0004900783333333333,
      "hi": 0.0004900783333333333,
      "n": 12,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "country=JP",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "country=JP",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "country=JP",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "country=MY",
      "target": "diy:claude-haiku-5.5",
      "value": 8460,
      "lo": 8460,
      "hi": 8460,
      "n": 369,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "country=MY",
      "target": "diy:claude-haiku-5.5",
      "value": 13255.399999999996,
      "lo": 13255.399999999996,
      "hi": 13255.399999999996,
      "n": 369,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "country=MY",
      "target": "diy:claude-haiku-5.5",
      "value": 16025.279999999999,
      "lo": 16025.279999999999,
      "hi": 16025.279999999999,
      "n": 369,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "country=MY",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0008723611111111104,
      "lo": 0.0008723611111111104,
      "hi": 0.0008723611111111104,
      "n": 369,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "country=MY",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0008723611111111104,
      "lo": 0.0008723611111111104,
      "hi": 0.0008723611111111104,
      "n": 369,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "country=MY",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.010303196519090837,
      "n": 369,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "country=MY",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.010303196519090837,
      "n": 369,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "country=MY",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.010303196519090837,
      "n": 369,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "currency=JPY",
      "target": "diy:claude-haiku-5.5",
      "value": 3595,
      "lo": 3595,
      "hi": 3595,
      "n": 12,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "currency=JPY",
      "target": "diy:claude-haiku-5.5",
      "value": 8550.5,
      "lo": 8550.5,
      "hi": 8550.5,
      "n": 12,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "currency=JPY",
      "target": "diy:claude-haiku-5.5",
      "value": 8959.7,
      "lo": 8959.7,
      "hi": 8959.7,
      "n": 12,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "currency=JPY",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0004900783333333333,
      "lo": 0.0004900783333333333,
      "hi": 0.0004900783333333333,
      "n": 12,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "currency=JPY",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0004900783333333333,
      "lo": 0.0004900783333333333,
      "hi": 0.0004900783333333333,
      "n": 12,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "currency=JPY",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "currency=JPY",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "currency=JPY",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "currency=MYR",
      "target": "diy:claude-haiku-5.5",
      "value": 8460,
      "lo": 8460,
      "hi": 8460,
      "n": 369,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "currency=MYR",
      "target": "diy:claude-haiku-5.5",
      "value": 13255.399999999996,
      "lo": 13255.399999999996,
      "hi": 13255.399999999996,
      "n": 369,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "currency=MYR",
      "target": "diy:claude-haiku-5.5",
      "value": 16025.279999999999,
      "lo": 16025.279999999999,
      "hi": 16025.279999999999,
      "n": 369,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "currency=MYR",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0008723611111111104,
      "lo": 0.0008723611111111104,
      "hi": 0.0008723611111111104,
      "n": 369,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "currency=MYR",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0008723611111111104,
      "lo": 0.0008723611111111104,
      "hi": 0.0008723611111111104,
      "n": 369,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "currency=MYR",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.010303196519090837,
      "n": 369,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "currency=MYR",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.010303196519090837,
      "n": 369,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "currency=MYR",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.010303196519090837,
      "n": 369,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "dateEra=old",
      "target": "diy:claude-haiku-5.5",
      "value": 8451,
      "lo": 8451,
      "hi": 8451,
      "n": 384,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "dateEra=old",
      "target": "diy:claude-haiku-5.5",
      "value": 13164.199999999993,
      "lo": 13164.199999999993,
      "hi": 13164.199999999993,
      "n": 384,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "dateEra=old",
      "target": "diy:claude-haiku-5.5",
      "value": 15983.430000000004,
      "lo": 15983.430000000004,
      "hi": 15983.430000000004,
      "n": 384,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "dateEra=old",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0008710539322916658,
      "lo": 0.0008710539322916658,
      "hi": 0.0008710539322916658,
      "n": 384,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "dateEra=old",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0008710539322916658,
      "lo": 0.0008710539322916658,
      "hi": 0.0008710539322916658,
      "n": 384,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "dateEra=old",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 8.673617379884035e-19,
      "hi": 0.009904714241677031,
      "n": 384,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "dateEra=old",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 8.673617379884035e-19,
      "hi": 0.009904714241677031,
      "n": 384,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "dateEra=old",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 8.673617379884035e-19,
      "hi": 0.009904714241677031,
      "n": 384,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "dateEra=recent",
      "target": "diy:claude-haiku-5.5",
      "value": 8214,
      "lo": 8214,
      "hi": 8214,
      "n": 69,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "dateEra=recent",
      "target": "diy:claude-haiku-5.5",
      "value": 13736.999999999995,
      "lo": 13736.999999999995,
      "hi": 13736.999999999995,
      "n": 69,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "dateEra=recent",
      "target": "diy:claude-haiku-5.5",
      "value": 15835.92,
      "lo": 15835.92,
      "hi": 15835.92,
      "n": 69,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "dateEra=recent",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0008596897101449274,
      "lo": 0.0008596897101449274,
      "hi": 0.0008596897101449274,
      "n": 69,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "dateEra=recent",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0008596897101449274,
      "lo": 0.0008596897101449274,
      "hi": 0.0008596897101449274,
      "n": 69,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "dateEra=recent",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 3.469446951953614e-18,
      "hi": 0.05273725818905173,
      "n": 69,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "dateEra=recent",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 3.469446951953614e-18,
      "hi": 0.05273725818905173,
      "n": 69,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "dateEra=recent",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 3.469446951953614e-18,
      "hi": 0.05273725818905173,
      "n": 69,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "docType=receipt",
      "target": "diy:claude-haiku-5.5",
      "value": 8424,
      "lo": 8424,
      "hi": 8424,
      "n": 453,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "docType=receipt",
      "target": "diy:claude-haiku-5.5",
      "value": 13234.599999999999,
      "lo": 13234.599999999999,
      "hi": 13234.599999999999,
      "n": 453,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "docType=receipt",
      "target": "diy:claude-haiku-5.5",
      "value": 15886.080000000002,
      "lo": 15886.080000000002,
      "hi": 15886.080000000002,
      "n": 453,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "docType=receipt",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0008693229580573947,
      "lo": 0.0008693229580573947,
      "hi": 0.0008693229580573947,
      "n": 453,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "docType=receipt",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0008693229580573947,
      "lo": 0.0008693229580573947,
      "hi": 0.0008693229580573947,
      "n": 453,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "docType=receipt",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.008408735123582253,
      "n": 453,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "docType=receipt",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.008408735123582253,
      "n": 453,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "docType=receipt",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.008408735123582253,
      "n": 453,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "editedField=date",
      "target": "diy:claude-haiku-5.5",
      "value": 8339,
      "lo": 8339,
      "hi": 8339,
      "n": 12,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "editedField=date",
      "target": "diy:claude-haiku-5.5",
      "value": 12986.75,
      "lo": 12986.75,
      "hi": 12986.75,
      "n": 12,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "editedField=date",
      "target": "diy:claude-haiku-5.5",
      "value": 13151.75,
      "lo": 13151.75,
      "hi": 13151.75,
      "n": 12,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "editedField=date",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0009238216666666667,
      "lo": 0.0009238216666666667,
      "hi": 0.0009238216666666667,
      "n": 12,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "editedField=date",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0009238216666666667,
      "lo": 0.0009238216666666667,
      "hi": 0.0009238216666666667,
      "n": 12,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "editedField=date",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "editedField=date",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "editedField=date",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "editedField=items+total",
      "target": "diy:claude-haiku-5.5",
      "value": 8843.5,
      "lo": 8843.5,
      "hi": 8843.5,
      "n": 12,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "editedField=items+total",
      "target": "diy:claude-haiku-5.5",
      "value": 12735.049999999997,
      "lo": 12735.049999999997,
      "hi": 12735.049999999997,
      "n": 12,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "editedField=items+total",
      "target": "diy:claude-haiku-5.5",
      "value": 13883.010000000002,
      "lo": 13883.010000000002,
      "hi": 13883.010000000002,
      "n": 12,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "editedField=items+total",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0008532591666666665,
      "lo": 0.0008532591666666665,
      "hi": 0.0008532591666666665,
      "n": 12,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "editedField=items+total",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0008532591666666665,
      "lo": 0.0008532591666666665,
      "hi": 0.0008532591666666665,
      "n": 12,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "editedField=items+total",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "editedField=items+total",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "editedField=items+total",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "editedField=total",
      "target": "diy:claude-haiku-5.5",
      "value": 9360,
      "lo": 9360,
      "hi": 9360,
      "n": 27,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "editedField=total",
      "target": "diy:claude-haiku-5.5",
      "value": 15324.199999999999,
      "lo": 15324.199999999999,
      "hi": 15324.199999999999,
      "n": 27,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "editedField=total",
      "target": "diy:claude-haiku-5.5",
      "value": 16064.98,
      "lo": 16064.98,
      "hi": 16064.98,
      "n": 27,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "editedField=total",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0011136303703703702,
      "lo": 0.0011136303703703702,
      "hi": 0.0011136303703703702,
      "n": 27,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "editedField=total",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0011136303703703702,
      "lo": 0.0011136303703703702,
      "hi": 0.0011136303703703702,
      "n": 27,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "editedField=total",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.12455502974186705,
      "n": 27,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "editedField=total",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.12455502974186705,
      "n": 27,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "editedField=total",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.12455502974186705,
      "n": 27,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "lang=en",
      "target": "diy:claude-haiku-5.5",
      "value": 8423,
      "lo": 8423,
      "hi": 8423,
      "n": 387,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "lang=en",
      "target": "diy:claude-haiku-5.5",
      "value": 13265.8,
      "lo": 13265.8,
      "hi": 13265.8,
      "n": 387,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "lang=en",
      "target": "diy:claude-haiku-5.5",
      "value": 15975.059999999996,
      "lo": 15975.059999999996,
      "hi": 15975.059999999996,
      "n": 387,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "lang=en",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0008703826098191207,
      "lo": 0.0008703826098191207,
      "hi": 0.0008703826098191207,
      "n": 387,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "lang=en",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0008703826098191207,
      "lo": 0.0008703826098191207,
      "hi": 0.0008703826098191207,
      "n": 387,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "lang=en",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.009828688164979105,
      "n": 387,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "lang=en",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.009828688164979105,
      "n": 387,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "lang=en",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.009828688164979105,
      "n": 387,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "lang=ja",
      "target": "diy:claude-haiku-5.5",
      "value": 3595,
      "lo": 3595,
      "hi": 3595,
      "n": 12,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "lang=ja",
      "target": "diy:claude-haiku-5.5",
      "value": 8550.5,
      "lo": 8550.5,
      "hi": 8550.5,
      "n": 12,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "lang=ja",
      "target": "diy:claude-haiku-5.5",
      "value": 8959.7,
      "lo": 8959.7,
      "hi": 8959.7,
      "n": 12,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "lang=ja",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0004900783333333333,
      "lo": 0.0004900783333333333,
      "hi": 0.0004900783333333333,
      "n": 12,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "lang=ja",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0004900783333333333,
      "lo": 0.0004900783333333333,
      "hi": 0.0004900783333333333,
      "n": 12,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "lang=ja",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "lang=ja",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "lang=ja",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "lang=zh",
      "target": "diy:claude-haiku-5.5",
      "value": 9301.5,
      "lo": 9301.5,
      "hi": 9301.5,
      "n": 18,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "lang=zh",
      "target": "diy:claude-haiku-5.5",
      "value": 13025.35,
      "lo": 13025.35,
      "hi": 13025.35,
      "n": 18,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "lang=zh",
      "target": "diy:claude-haiku-5.5",
      "value": 13085.869999999999,
      "lo": 13085.869999999999,
      "hi": 13085.869999999999,
      "n": 18,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "lang=zh",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0008684411111111109,
      "lo": 0.0008684411111111109,
      "hi": 0.0008684411111111109,
      "n": 18,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "lang=zh",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0008684411111111109,
      "lo": 0.0008684411111111109,
      "hi": 0.0008684411111111109,
      "n": 18,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "lang=zh",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.17587922364665765,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "lang=zh",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.17587922364665765,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "lang=zh",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.17587922364665765,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "prompt=explicit_amounts",
      "target": "diy:claude-haiku-5.5",
      "value": 8311.5,
      "lo": 8311.5,
      "hi": 8311.5,
      "n": 24,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "prompt=explicit_amounts",
      "target": "diy:claude-haiku-5.5",
      "value": 11707.149999999998,
      "lo": 11707.149999999998,
      "hi": 11707.149999999998,
      "n": 24,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "prompt=explicit_amounts",
      "target": "diy:claude-haiku-5.5",
      "value": 11908.72,
      "lo": 11908.72,
      "hi": 11908.72,
      "n": 24,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "prompt=explicit_amounts",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0008542445833333332,
      "lo": 0.0008542445833333332,
      "hi": 0.0008542445833333332,
      "n": 24,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "prompt=explicit_amounts",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0008542445833333332,
      "lo": 0.0008542445833333332,
      "hi": 0.0008542445833333332,
      "n": 24,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "prompt=explicit_amounts",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 1.3877787807814457e-17,
      "hi": 0.13797620467498017,
      "n": 24,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "prompt=explicit_amounts",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 1.3877787807814457e-17,
      "hi": 0.13797620467498017,
      "n": 24,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "prompt=explicit_amounts",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 1.3877787807814457e-17,
      "hi": 0.13797620467498017,
      "n": 24,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "prompt=free",
      "target": "diy:claude-haiku-5.5",
      "value": 8214,
      "lo": 8214,
      "hi": 8214,
      "n": 15,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "prompt=free",
      "target": "diy:claude-haiku-5.5",
      "value": 11311.6,
      "lo": 11311.6,
      "hi": 11311.6,
      "n": 15,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "prompt=free",
      "target": "diy:claude-haiku-5.5",
      "value": 11559.119999999999,
      "lo": 11559.119999999999,
      "hi": 11559.119999999999,
      "n": 15,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "prompt=free",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0008842306666666665,
      "lo": 0.0008842306666666665,
      "hi": 0.0008842306666666665,
      "n": 15,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "prompt=free",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0008842306666666665,
      "lo": 0.0008842306666666665,
      "hi": 0.0008842306666666665,
      "n": 15,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "prompt=free",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "prompt=free",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "prompt=free",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "prompt=target_total",
      "target": "diy:claude-haiku-5.5",
      "value": 10072,
      "lo": 10072,
      "hi": 10072,
      "n": 21,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "prompt=target_total",
      "target": "diy:claude-haiku-5.5",
      "value": 12021,
      "lo": 12021,
      "hi": 12021,
      "n": 21,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "prompt=target_total",
      "target": "diy:claude-haiku-5.5",
      "value": 14510.600000000002,
      "lo": 14510.600000000002,
      "hi": 14510.600000000002,
      "n": 21,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "prompt=target_total",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0010411276190476187,
      "lo": 0.0010411276190476187,
      "hi": 0.0010411276190476187,
      "n": 21,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "prompt=target_total",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0010411276190476187,
      "lo": 0.0010411276190476187,
      "hi": 0.0010411276190476187,
      "n": 21,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "prompt=target_total",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.154639018924847,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "prompt=target_total",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.154639018924847,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "prompt=target_total",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.154639018924847,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "provenance=intact",
      "target": "diy:claude-haiku-5.5",
      "value": 9094,
      "lo": 9094,
      "hi": 9094,
      "n": 60,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "provenance=intact",
      "target": "diy:claude-haiku-5.5",
      "value": 14012.3,
      "lo": 14012.3,
      "hi": 14012.3,
      "n": 60,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "provenance=intact",
      "target": "diy:claude-haiku-5.5",
      "value": 15257.429999999995,
      "lo": 15257.429999999995,
      "hi": 15257.429999999995,
      "n": 60,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "provenance=intact",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0009559188333333335,
      "lo": 0.0009559188333333335,
      "hi": 0.0009559188333333335,
      "n": 60,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "provenance=intact",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0009559188333333335,
      "lo": 0.0009559188333333335,
      "hi": 0.0009559188333333335,
      "n": 60,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "provenance=intact",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.06017185214208986,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "provenance=intact",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.06017185214208986,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "provenance=intact",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.06017185214208986,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "provenance=none",
      "target": "diy:claude-haiku-5.5",
      "value": 8133,
      "lo": 8133,
      "hi": 8133,
      "n": 333,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "provenance=none",
      "target": "diy:claude-haiku-5.5",
      "value": 12907.599999999997,
      "lo": 12907.599999999997,
      "hi": 12907.599999999997,
      "n": 333,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "provenance=none",
      "target": "diy:claude-haiku-5.5",
      "value": 15905.28,
      "lo": 15905.28,
      "hi": 15905.28,
      "n": 333,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "provenance=none",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0008352761561561555,
      "lo": 0.0008352761561561555,
      "hi": 0.0008352761561561555,
      "n": 333,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "provenance=none",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0008352761561561555,
      "lo": 0.0008352761561561555,
      "hi": 0.0008352761561561555,
      "n": 333,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "provenance=none",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.011404352760326315,
      "n": 333,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "provenance=none",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.011404352760326315,
      "n": 333,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "provenance=none",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.011404352760326315,
      "n": 333,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "provenance=stripped",
      "target": "diy:claude-haiku-5.5",
      "value": 8935.5,
      "lo": 8935.5,
      "hi": 8935.5,
      "n": 60,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "provenance=stripped",
      "target": "diy:claude-haiku-5.5",
      "value": 13674.649999999998,
      "lo": 13674.649999999998,
      "hi": 13674.649999999998,
      "n": 60,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "provenance=stripped",
      "target": "diy:claude-haiku-5.5",
      "value": 15340.049999999997,
      "lo": 15340.049999999997,
      "hi": 15340.049999999997,
      "n": 60,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "provenance=stripped",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0009716868333333335,
      "lo": 0.0009716868333333335,
      "hi": 0.0009716868333333335,
      "n": 60,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "provenance=stripped",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0009716868333333335,
      "lo": 0.0009716868333333335,
      "hi": 0.0009716868333333335,
      "n": 60,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "provenance=stripped",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.06017185214208986,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "provenance=stripped",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.06017185214208986,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "provenance=stripped",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.06017185214208986,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "source=commons",
      "target": "diy:claude-haiku-5.5",
      "value": 7946,
      "lo": 7946,
      "hi": 7946,
      "n": 39,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "source=commons",
      "target": "diy:claude-haiku-5.5",
      "value": 14328.300000000003,
      "lo": 14328.300000000003,
      "hi": 14328.300000000003,
      "n": 39,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "source=commons",
      "target": "diy:claude-haiku-5.5",
      "value": 15837.72,
      "lo": 15837.72,
      "hi": 15837.72,
      "n": 39,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "source=commons",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0008019474358974358,
      "lo": 0.0008019474358974358,
      "hi": 0.0008019474358974358,
      "n": 39,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "source=commons",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0008019474358974358,
      "lo": 0.0008019474358974358,
      "hi": 0.0008019474358974358,
      "n": 39,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "source=commons",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.08966685370757144,
      "n": 39,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "source=commons",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.08966685370757144,
      "n": 39,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "source=commons",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.08966685370757144,
      "n": 39,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "source=edited:gemini-nano-banana-2.1",
      "target": "diy:claude-haiku-5.5",
      "value": 8094.5,
      "lo": 8094.5,
      "hi": 8094.5,
      "n": 12,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "source=edited:gemini-nano-banana-2.1",
      "target": "diy:claude-haiku-5.5",
      "value": 10369.499999999998,
      "lo": 10369.499999999998,
      "hi": 10369.499999999998,
      "n": 12,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "source=edited:gemini-nano-banana-2.1",
      "target": "diy:claude-haiku-5.5",
      "value": 11614.7,
      "lo": 11614.7,
      "hi": 11614.7,
      "n": 12,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "source=edited:gemini-nano-banana-2.1",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0008319608333333333,
      "lo": 0.0008319608333333333,
      "hi": 0.0008319608333333333,
      "n": 12,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "source=edited:gemini-nano-banana-2.1",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0008319608333333333,
      "lo": 0.0008319608333333333,
      "hi": 0.0008319608333333333,
      "n": 12,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "source=edited:gemini-nano-banana-2.1",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "source=edited:gemini-nano-banana-2.1",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "source=edited:gemini-nano-banana-2.1",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "source=edited:gpt-image-2.5-sunburst",
      "target": "diy:claude-haiku-5.5",
      "value": 6403,
      "lo": 6403,
      "hi": 6403,
      "n": 12,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "source=edited:gpt-image-2.5-sunburst",
      "target": "diy:claude-haiku-5.5",
      "value": 9352.85,
      "lo": 9352.85,
      "hi": 9352.85,
      "n": 12,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "source=edited:gpt-image-2.5-sunburst",
      "target": "diy:claude-haiku-5.5",
      "value": 9358.57,
      "lo": 9358.57,
      "hi": 9358.57,
      "n": 12,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "source=edited:gpt-image-2.5-sunburst",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0006404925000000002,
      "lo": 0.0006404925000000002,
      "hi": 0.0006404925000000002,
      "n": 12,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "source=edited:gpt-image-2.5-sunburst",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0006404925000000002,
      "lo": 0.0006404925000000002,
      "hi": 0.0006404925000000002,
      "n": 12,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "source=edited:gpt-image-2.5-sunburst",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "source=edited:gpt-image-2.5-sunburst",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "source=edited:gpt-image-2.5-sunburst",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "source=edited:sharp-patch",
      "target": "diy:claude-haiku-5.5",
      "value": 11458.5,
      "lo": 11458.5,
      "hi": 11458.5,
      "n": 30,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "source=edited:sharp-patch",
      "target": "diy:claude-haiku-5.5",
      "value": 15167.299999999997,
      "lo": 15167.299999999997,
      "hi": 15167.299999999997,
      "n": 30,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "source=edited:sharp-patch",
      "target": "diy:claude-haiku-5.5",
      "value": 16047.67,
      "lo": 16047.67,
      "hi": 16047.67,
      "n": 30,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "source=edited:sharp-patch",
      "target": "diy:claude-haiku-5.5",
      "value": 0.001195770333333333,
      "lo": 0.001195770333333333,
      "hi": 0.001195770333333333,
      "n": 30,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "source=edited:sharp-patch",
      "target": "diy:claude-haiku-5.5",
      "value": 0.001195770333333333,
      "lo": 0.001195770333333333,
      "hi": 0.001195770333333333,
      "n": 30,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "source=edited:sharp-patch",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "source=edited:sharp-patch",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "source=edited:sharp-patch",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "source=generated:flux-3-image",
      "target": "diy:claude-haiku-5.5",
      "value": 9416,
      "lo": 9416,
      "hi": 9416,
      "n": 18,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "source=generated:flux-3-image",
      "target": "diy:claude-haiku-5.5",
      "value": 11902,
      "lo": 11902,
      "hi": 11902,
      "n": 18,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "source=generated:flux-3-image",
      "target": "diy:claude-haiku-5.5",
      "value": 11997.199999999999,
      "lo": 11997.199999999999,
      "hi": 11997.199999999999,
      "n": 18,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "source=generated:flux-3-image",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0010190644444444442,
      "lo": 0.0010190644444444442,
      "hi": 0.0010190644444444442,
      "n": 18,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "source=generated:flux-3-image",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0010190644444444442,
      "lo": 0.0010190644444444442,
      "hi": 0.0010190644444444442,
      "n": 18,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "source=generated:flux-3-image",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.17587922364665765,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "source=generated:flux-3-image",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.17587922364665765,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "source=generated:flux-3-image",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.17587922364665765,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "source=generated:gemini-nano-banana-2.1",
      "target": "diy:claude-haiku-5.5",
      "value": 9522,
      "lo": 9522,
      "hi": 9522,
      "n": 21,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "source=generated:gemini-nano-banana-2.1",
      "target": "diy:claude-haiku-5.5",
      "value": 11917,
      "lo": 11917,
      "hi": 11917,
      "n": 21,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "source=generated:gemini-nano-banana-2.1",
      "target": "diy:claude-haiku-5.5",
      "value": 14489.800000000003,
      "lo": 14489.800000000003,
      "hi": 14489.800000000003,
      "n": 21,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "source=generated:gemini-nano-banana-2.1",
      "target": "diy:claude-haiku-5.5",
      "value": 0.000980927619047619,
      "lo": 0.000980927619047619,
      "hi": 0.000980927619047619,
      "n": 21,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "source=generated:gemini-nano-banana-2.1",
      "target": "diy:claude-haiku-5.5",
      "value": 0.000980927619047619,
      "lo": 0.000980927619047619,
      "hi": 0.000980927619047619,
      "n": 21,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "source=generated:gemini-nano-banana-2.1",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.154639018924847,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "source=generated:gemini-nano-banana-2.1",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.154639018924847,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "source=generated:gemini-nano-banana-2.1",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.154639018924847,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "source=generated:gpt-image-2.5-flare",
      "target": "diy:claude-haiku-5.5",
      "value": 7894,
      "lo": 7894,
      "hi": 7894,
      "n": 21,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "source=generated:gpt-image-2.5-flare",
      "target": "diy:claude-haiku-5.5",
      "value": 10806,
      "lo": 10806,
      "hi": 10806,
      "n": 21,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "source=generated:gpt-image-2.5-flare",
      "target": "diy:claude-haiku-5.5",
      "value": 10884.4,
      "lo": 10884.4,
      "hi": 10884.4,
      "n": 21,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "source=generated:gpt-image-2.5-flare",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0007945890476190478,
      "lo": 0.0007945890476190478,
      "hi": 0.0007945890476190478,
      "n": 21,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "source=generated:gpt-image-2.5-flare",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0007945890476190478,
      "lo": 0.0007945890476190478,
      "hi": 0.0007945890476190478,
      "n": 21,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "source=generated:gpt-image-2.5-flare",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.154639018924847,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "source=generated:gpt-image-2.5-flare",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.154639018924847,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "source=generated:gpt-image-2.5-flare",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.154639018924847,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "source=sroie",
      "target": "diy:claude-haiku-5.5",
      "value": 8257,
      "lo": 8257,
      "hi": 8257,
      "n": 294,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "source=sroie",
      "target": "diy:claude-haiku-5.5",
      "value": 12688.099999999997,
      "lo": 12688.099999999997,
      "hi": 12688.099999999997,
      "n": 294,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "source=sroie",
      "target": "diy:claude-haiku-5.5",
      "value": 15993.889999999994,
      "lo": 15993.889999999994,
      "hi": 15993.889999999994,
      "n": 294,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "source=sroie",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0008396973129251699,
      "lo": 0.0008396973129251699,
      "hi": 0.0008396973129251699,
      "n": 294,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "source=sroie",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0008396973129251699,
      "lo": 0.0008396973129251699,
      "hi": 0.0008396973129251699,
      "n": 294,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "source=sroie",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.012897663192708011,
      "n": 294,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "source=sroie",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.012897663192708011,
      "n": 294,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "source=sroie",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.012897663192708011,
      "n": 294,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "transform=crop",
      "target": "diy:claude-haiku-5.5",
      "value": 9195,
      "lo": 9195,
      "hi": 9195,
      "n": 15,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "transform=crop",
      "target": "diy:claude-haiku-5.5",
      "value": 13881.499999999996,
      "lo": 13881.499999999996,
      "hi": 13881.499999999996,
      "n": 15,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "transform=crop",
      "target": "diy:claude-haiku-5.5",
      "value": 15525.099999999999,
      "lo": 15525.099999999999,
      "hi": 15525.099999999999,
      "n": 15,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "transform=crop",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0009543966666666667,
      "lo": 0.0009543966666666667,
      "hi": 0.0009543966666666667,
      "n": 15,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "transform=crop",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0009543966666666667,
      "lo": 0.0009543966666666667,
      "hi": 0.0009543966666666667,
      "n": 15,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "transform=crop",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "transform=crop",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "transform=crop",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "transform=jpeg",
      "target": "diy:claude-haiku-5.5",
      "value": 7365,
      "lo": 7365,
      "hi": 7365,
      "n": 15,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "transform=jpeg",
      "target": "diy:claude-haiku-5.5",
      "value": 11360.099999999999,
      "lo": 11360.099999999999,
      "hi": 11360.099999999999,
      "n": 15,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "transform=jpeg",
      "target": "diy:claude-haiku-5.5",
      "value": 12310.419999999998,
      "lo": 12310.419999999998,
      "hi": 12310.419999999998,
      "n": 15,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "transform=jpeg",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0008204760000000001,
      "lo": 0.0008204760000000001,
      "hi": 0.0008204760000000001,
      "n": 15,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "transform=jpeg",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0008204760000000001,
      "lo": 0.0008204760000000001,
      "hi": 0.0008204760000000001,
      "n": 15,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "transform=jpeg",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "transform=jpeg",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "transform=jpeg",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "transform=rephoto",
      "target": "diy:claude-haiku-5.5",
      "value": 9467,
      "lo": 9467,
      "hi": 9467,
      "n": 15,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "transform=rephoto",
      "target": "diy:claude-haiku-5.5",
      "value": 11950.3,
      "lo": 11950.3,
      "hi": 11950.3,
      "n": 15,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "transform=rephoto",
      "target": "diy:claude-haiku-5.5",
      "value": 12785.259999999998,
      "lo": 12785.259999999998,
      "hi": 12785.259999999998,
      "n": 15,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "transform=rephoto",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0008844139999999998,
      "lo": 0.0008844139999999998,
      "hi": 0.0008844139999999998,
      "n": 15,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "transform=rephoto",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0008844139999999998,
      "lo": 0.0008844139999999998,
      "hi": 0.0008844139999999998,
      "n": 15,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "transform=rephoto",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "transform=rephoto",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "transform=rephoto",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "transform=resize",
      "target": "diy:claude-haiku-5.5",
      "value": 8364,
      "lo": 8364,
      "hi": 8364,
      "n": 15,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "transform=resize",
      "target": "diy:claude-haiku-5.5",
      "value": 9721.8,
      "lo": 9721.8,
      "hi": 9721.8,
      "n": 15,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "transform=resize",
      "target": "diy:claude-haiku-5.5",
      "value": 10229.16,
      "lo": 10229.16,
      "hi": 10229.16,
      "n": 15,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "transform=resize",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0008212986666666667,
      "lo": 0.0008212986666666667,
      "hi": 0.0008212986666666667,
      "n": 15,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "transform=resize",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0008212986666666667,
      "lo": 0.0008212986666666667,
      "hi": 0.0008212986666666667,
      "n": 15,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "transform=resize",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "transform=resize",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "transform=resize",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "transform=rotate",
      "target": "diy:claude-haiku-5.5",
      "value": 8313,
      "lo": 8313,
      "hi": 8313,
      "n": 15,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "transform=rotate",
      "target": "diy:claude-haiku-5.5",
      "value": 11146.4,
      "lo": 11146.4,
      "hi": 11146.4,
      "n": 15,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "transform=rotate",
      "target": "diy:claude-haiku-5.5",
      "value": 11514.88,
      "lo": 11514.88,
      "hi": 11514.88,
      "n": 15,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "transform=rotate",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0007585126666666668,
      "lo": 0.0007585126666666668,
      "hi": 0.0007585126666666668,
      "n": 15,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "transform=rotate",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0007585126666666668,
      "lo": 0.0007585126666666668,
      "hi": 0.0007585126666666668,
      "n": 15,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "transform=rotate",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "transform=rotate",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "transform=rotate",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "transform=screenshot",
      "target": "diy:claude-haiku-5.5",
      "value": 7528,
      "lo": 7528,
      "hi": 7528,
      "n": 15,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "transform=screenshot",
      "target": "diy:claude-haiku-5.5",
      "value": 11765.099999999997,
      "lo": 11765.099999999997,
      "hi": 11765.099999999997,
      "n": 15,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "transform=screenshot",
      "target": "diy:claude-haiku-5.5",
      "value": 13118.619999999999,
      "lo": 13118.619999999999,
      "hi": 13118.619999999999,
      "n": 15,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "transform=screenshot",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0008422473333333332,
      "lo": 0.0008422473333333332,
      "hi": 0.0008422473333333332,
      "n": 15,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "transform=screenshot",
      "target": "diy:claude-haiku-5.5",
      "value": 0.0008422473333333332,
      "lo": 0.0008422473333333332,
      "hi": 0.0008422473333333332,
      "n": 15,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "transform=screenshot",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "transform=screenshot",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "transform=screenshot",
      "target": "diy:claude-haiku-5.5",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "all",
      "target": "diy:gemini-3.8-flash",
      "value": 0.25964912280701735,
      "lo": 0.18205180924999986,
      "hi": 0.35592883964441313,
      "n": 95,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "all",
      "target": "diy:gemini-3.8-flash",
      "value": 0.2619047619047616,
      "lo": 0.1893467213237418,
      "hi": 0.35025392013670703,
      "n": 112,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "all",
      "target": "diy:gemini-3.8-flash",
      "value": 0.26515151515151497,
      "lo": 0.18416519784516566,
      "hi": 0.36578387892167735,
      "n": 88,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.subtotal",
      "stratum": "all",
      "target": "diy:gemini-3.8-flash",
      "value": 0.22916666666666669,
      "lo": 0.08937790137618432,
      "hi": 0.4738262589133022,
      "n": 16,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.tax",
      "stratum": "all",
      "target": "diy:gemini-3.8-flash",
      "value": 0.28333333333333327,
      "lo": 0.1340262550429334,
      "hi": 0.5024613134466216,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "all",
      "target": "diy:gemini-3.8-flash",
      "value": 0.2699724517906332,
      "lo": 0.19885456121008174,
      "hi": 0.3552465587822836,
      "n": 121,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.receipt_number",
      "stratum": "all",
      "target": "diy:gemini-3.8-flash",
      "value": 0.2727272727272727,
      "lo": 0.09746059290024889,
      "hi": 0.5656453011761291,
      "n": 11,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "all",
      "target": "diy:gemini-3.8-flash",
      "value": 0.25895316804407675,
      "lo": 0.18917059682912157,
      "hi": 0.3435700977736565,
      "n": 121,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "line_items_f1",
      "stratum": "all",
      "target": "diy:gemini-3.8-flash",
      "value": 0.4036144578313253,
      "lo": 0.3124953078078078,
      "hi": 0.47478484471401095,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "all",
      "target": "diy:gemini-3.8-flash",
      "value": 0.2699724517906332,
      "lo": 0.19885456121008174,
      "hi": 0.3552465587822836,
      "n": 121,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "arithmetic=consistent",
      "target": "diy:gemini-3.8-flash",
      "value": 0.29333333333333317,
      "lo": 0.20240140321903238,
      "hi": 0.40440445167386047,
      "n": 75,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "arithmetic=consistent",
      "target": "diy:gemini-3.8-flash",
      "value": 0.2934782608695649,
      "lo": 0.21021716711066368,
      "hi": 0.3932947110725273,
      "n": 92,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "arithmetic=consistent",
      "target": "diy:gemini-3.8-flash",
      "value": 0.3013698630136985,
      "lo": 0.2082259057912606,
      "hi": 0.4143736590130706,
      "n": 73,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.subtotal",
      "stratum": "arithmetic=consistent",
      "target": "diy:gemini-3.8-flash",
      "value": 0.2424242424242424,
      "lo": 0.08110064930098895,
      "hi": 0.5370860306522395,
      "n": 11,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.tax",
      "stratum": "arithmetic=consistent",
      "target": "diy:gemini-3.8-flash",
      "value": 0.3076923076923077,
      "lo": 0.12680703655710515,
      "hi": 0.5763065681945096,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "arithmetic=consistent",
      "target": "diy:gemini-3.8-flash",
      "value": 0.2999999999999996,
      "lo": 0.2189488529493272,
      "hi": 0.3958485463334662,
      "n": 100,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "arithmetic=consistent",
      "target": "diy:gemini-3.8-flash",
      "value": 0.28666666666666635,
      "lo": 0.20722567437315018,
      "hi": 0.3818915515284963,
      "n": 100,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "line_items_f1",
      "stratum": "arithmetic=consistent",
      "target": "diy:gemini-3.8-flash",
      "value": 0.4343891402714932,
      "lo": 0.33043478260869563,
      "hi": 0.5,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "arithmetic=consistent",
      "target": "diy:gemini-3.8-flash",
      "value": 0.2999999999999996,
      "lo": 0.2189488529493272,
      "hi": 0.3958485463334662,
      "n": 100,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "arithmetic=inconsistent",
      "target": "diy:gemini-3.8-flash",
      "value": 0.0980392156862745,
      "lo": 0.02452053761902165,
      "hi": 0.31973521797632976,
      "n": 17,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "arithmetic=inconsistent",
      "target": "diy:gemini-3.8-flash",
      "value": 0.09259259259259259,
      "lo": 0.02313862362313196,
      "hi": 0.305355558607478,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "arithmetic=inconsistent",
      "target": "diy:gemini-3.8-flash",
      "value": 0.07142857142857142,
      "lo": 0.012722215247890939,
      "hi": 0.31468704424151117,
      "n": 14,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "arithmetic=inconsistent",
      "target": "diy:gemini-3.8-flash",
      "value": 0.09259259259259259,
      "lo": 0.02313862362313196,
      "hi": 0.305355558607478,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "arithmetic=inconsistent",
      "target": "diy:gemini-3.8-flash",
      "value": 0.09259259259259259,
      "lo": 0.02313862362313196,
      "hi": 0.305355558607478,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "arithmetic=inconsistent",
      "target": "diy:gemini-3.8-flash",
      "value": 0.09259259259259259,
      "lo": 0.02313862362313196,
      "hi": 0.305355558607478,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "base=sroie",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.1843181350088518,
      "n": 17,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "base=sroie",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.16112515805281938,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "base=sroie",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.1936076805344365,
      "n": 16,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "base=sroie",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.16112515805281938,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "base=sroie",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.16112515805281938,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "base=sroie",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.16112515805281938,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "capture=photo",
      "target": "diy:gemini-3.8-flash",
      "value": 0.2857142857142857,
      "lo": 0.11721378644576919,
      "hi": 0.5464908432989318,
      "n": 14,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "capture=photo",
      "target": "diy:gemini-3.8-flash",
      "value": 0.3015873015873015,
      "lo": 0.14920300804175174,
      "hi": 0.5153362851823937,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "capture=photo",
      "target": "diy:gemini-3.8-flash",
      "value": 0.3076923076923077,
      "lo": 0.12680703655710515,
      "hi": 0.5763065681945096,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.tax",
      "stratum": "capture=photo",
      "target": "diy:gemini-3.8-flash",
      "value": 0.2777777777777778,
      "lo": 0.10461389175889282,
      "hi": 0.5587167778656588,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "capture=photo",
      "target": "diy:gemini-3.8-flash",
      "value": 0.3015873015873015,
      "lo": 0.14920300804175174,
      "hi": 0.5153362851823937,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "capture=photo",
      "target": "diy:gemini-3.8-flash",
      "value": 0.3015873015873015,
      "lo": 0.14920300804175174,
      "hi": 0.5153362851823937,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "line_items_f1",
      "stratum": "capture=photo",
      "target": "diy:gemini-3.8-flash",
      "value": 0.39800995024875624,
      "lo": 0.2627722007722009,
      "hi": 0.5,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "capture=photo",
      "target": "diy:gemini-3.8-flash",
      "value": 0.3015873015873015,
      "lo": 0.14920300804175174,
      "hi": 0.5153362851823937,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "capture=scan",
      "target": "diy:gemini-3.8-flash",
      "value": 0.25514403292181065,
      "lo": 0.17280788384469353,
      "hi": 0.3596534007909335,
      "n": 81,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "capture=scan",
      "target": "diy:gemini-3.8-flash",
      "value": 0.2518518518518517,
      "lo": 0.17359965938733746,
      "hi": 0.3504202436705133,
      "n": 90,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "capture=scan",
      "target": "diy:gemini-3.8-flash",
      "value": 0.25675675675675674,
      "lo": 0.17097707299565887,
      "hi": 0.36654444130528563,
      "n": 74,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "capture=scan",
      "target": "diy:gemini-3.8-flash",
      "value": 0.2626262626262624,
      "lo": 0.18598149683086684,
      "hi": 0.357004371524566,
      "n": 99,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "capture=scan",
      "target": "diy:gemini-3.8-flash",
      "value": 0.24915824915824897,
      "lo": 0.1744103868950867,
      "hi": 0.34264560165072544,
      "n": 99,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "capture=scan",
      "target": "diy:gemini-3.8-flash",
      "value": 0.2626262626262624,
      "lo": 0.18598149683086684,
      "hi": 0.357004371524566,
      "n": 99,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "class=ai_full",
      "target": "diy:gemini-3.8-flash",
      "value": 0.28070175438596484,
      "lo": 0.12965494260365815,
      "hi": 0.5055113837352085,
      "n": 19,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "class=ai_full",
      "target": "diy:gemini-3.8-flash",
      "value": 0.28333333333333327,
      "lo": 0.1340262550429334,
      "hi": 0.5024613134466216,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "class=ai_full",
      "target": "diy:gemini-3.8-flash",
      "value": 0.2777777777777778,
      "lo": 0.10461389175889282,
      "hi": 0.5587167778656588,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.subtotal",
      "stratum": "class=ai_full",
      "target": "diy:gemini-3.8-flash",
      "value": 0.22916666666666669,
      "lo": 0.08937790137618432,
      "hi": 0.4738262589133022,
      "n": 16,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.tax",
      "stratum": "class=ai_full",
      "target": "diy:gemini-3.8-flash",
      "value": 0.28333333333333327,
      "lo": 0.1340262550429334,
      "hi": 0.5024613134466216,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "class=ai_full",
      "target": "diy:gemini-3.8-flash",
      "value": 0.28333333333333327,
      "lo": 0.1340262550429334,
      "hi": 0.5024613134466216,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.receipt_number",
      "stratum": "class=ai_full",
      "target": "diy:gemini-3.8-flash",
      "value": 0.2727272727272727,
      "lo": 0.09746059290024889,
      "hi": 0.5656453011761291,
      "n": 11,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "class=ai_full",
      "target": "diy:gemini-3.8-flash",
      "value": 0.24999999999999994,
      "lo": 0.11186170140766563,
      "hi": 0.468700877618744,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "line_items_f1",
      "stratum": "class=ai_full",
      "target": "diy:gemini-3.8-flash",
      "value": 0.4036144578313253,
      "lo": 0.3130394771939184,
      "hi": 0.47273543821448005,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "class=ai_full",
      "target": "diy:gemini-3.8-flash",
      "value": 0.28333333333333327,
      "lo": 0.1340262550429334,
      "hi": 0.5024613134466216,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "class=ai_inpaint",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.2775327998628892,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "class=ai_inpaint",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.2775327998628892,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "class=ai_inpaint",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.2775327998628892,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "class=ai_inpaint",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.2775327998628892,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "class=ai_inpaint",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.2775327998628892,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "class=ai_inpaint",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.2775327998628892,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "class=genuine",
      "target": "diy:gemini-3.8-flash",
      "value": 0.32653061224489816,
      "lo": 0.21207497001811879,
      "hi": 0.46620794685429856,
      "n": 49,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "class=genuine",
      "target": "diy:gemini-3.8-flash",
      "value": 0.32795698924731176,
      "lo": 0.2241533031833493,
      "hi": 0.45183605495905826,
      "n": 62,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "class=genuine",
      "target": "diy:gemini-3.8-flash",
      "value": 0.3333333333333335,
      "lo": 0.21969931853588529,
      "hi": 0.47031622143029206,
      "n": 51,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "class=genuine",
      "target": "diy:gemini-3.8-flash",
      "value": 0.3333333333333331,
      "lo": 0.23413420812686653,
      "hi": 0.449873479343222,
      "n": 70,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "class=genuine",
      "target": "diy:gemini-3.8-flash",
      "value": 0.3238095238095236,
      "lo": 0.22585488569859297,
      "hi": 0.4400960981983579,
      "n": 70,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "class=genuine",
      "target": "diy:gemini-3.8-flash",
      "value": 0.3333333333333331,
      "lo": 0.23413420812686653,
      "hi": 0.449873479343222,
      "n": 70,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "class=hard_negative",
      "target": "diy:gemini-3.8-flash",
      "value": 0.33333333333333337,
      "lo": 0.1269754121301785,
      "hi": 0.632202187824118,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "class=hard_negative",
      "target": "diy:gemini-3.8-flash",
      "value": 0.33333333333333337,
      "lo": 0.1269754121301785,
      "hi": 0.632202187824118,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "class=hard_negative",
      "target": "diy:gemini-3.8-flash",
      "value": 0.33333333333333337,
      "lo": 0.132793283187926,
      "hi": 0.6201510391347513,
      "n": 11,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "class=hard_negative",
      "target": "diy:gemini-3.8-flash",
      "value": 0.33333333333333337,
      "lo": 0.132793283187926,
      "hi": 0.6201510391347513,
      "n": 11,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "class=hard_negative",
      "target": "diy:gemini-3.8-flash",
      "value": 0.33333333333333337,
      "lo": 0.132793283187926,
      "hi": 0.6201510391347513,
      "n": 11,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "class=manual_edit",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.2775327998628892,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "class=manual_edit",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.2775327998628892,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "class=manual_edit",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.2775327998628892,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "class=manual_edit",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.2775327998628892,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "country=MY",
      "target": "diy:gemini-3.8-flash",
      "value": 0.2510822510822511,
      "lo": 0.1676462276761999,
      "hi": 0.3581746333951416,
      "n": 77,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "country=MY",
      "target": "diy:gemini-3.8-flash",
      "value": 0.24603174603174596,
      "lo": 0.16638803865952473,
      "hi": 0.3478883948209373,
      "n": 84,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "country=MY",
      "target": "diy:gemini-3.8-flash",
      "value": 0.25352112676056343,
      "lo": 0.16680551163187118,
      "hi": 0.3655392535356101,
      "n": 71,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "country=MY",
      "target": "diy:gemini-3.8-flash",
      "value": 0.2580645161290321,
      "lo": 0.17998522171036926,
      "hi": 0.3553377634243411,
      "n": 93,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "country=MY",
      "target": "diy:gemini-3.8-flash",
      "value": 0.2437275985663081,
      "lo": 0.16778270450312113,
      "hi": 0.3400038649358683,
      "n": 93,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "country=MY",
      "target": "diy:gemini-3.8-flash",
      "value": 0.2580645161290321,
      "lo": 0.17998522171036926,
      "hi": 0.3553377634243411,
      "n": 93,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "currency=MYR",
      "target": "diy:gemini-3.8-flash",
      "value": 0.2510822510822511,
      "lo": 0.1676462276761999,
      "hi": 0.3581746333951416,
      "n": 77,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "currency=MYR",
      "target": "diy:gemini-3.8-flash",
      "value": 0.24603174603174596,
      "lo": 0.16638803865952473,
      "hi": 0.3478883948209373,
      "n": 84,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "currency=MYR",
      "target": "diy:gemini-3.8-flash",
      "value": 0.25352112676056343,
      "lo": 0.16680551163187118,
      "hi": 0.3655392535356101,
      "n": 71,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "currency=MYR",
      "target": "diy:gemini-3.8-flash",
      "value": 0.2580645161290321,
      "lo": 0.17998522171036926,
      "hi": 0.3553377634243411,
      "n": 93,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "currency=MYR",
      "target": "diy:gemini-3.8-flash",
      "value": 0.2437275985663081,
      "lo": 0.16778270450312113,
      "hi": 0.3400038649358683,
      "n": 93,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "currency=MYR",
      "target": "diy:gemini-3.8-flash",
      "value": 0.2580645161290321,
      "lo": 0.17998522171036926,
      "hi": 0.3553377634243411,
      "n": 93,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "dateEra=old",
      "target": "diy:gemini-3.8-flash",
      "value": 0.25609756097560965,
      "lo": 0.1740357615326262,
      "hi": 0.3599889329503494,
      "n": 82,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "dateEra=old",
      "target": "diy:gemini-3.8-flash",
      "value": 0.2509363295880149,
      "lo": 0.1724518699542078,
      "hi": 0.3500315776728391,
      "n": 89,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "dateEra=old",
      "target": "diy:gemini-3.8-flash",
      "value": 0.25675675675675674,
      "lo": 0.17097707299565887,
      "hi": 0.36654444130528563,
      "n": 74,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.tax",
      "stratum": "dateEra=old",
      "target": "diy:gemini-3.8-flash",
      "value": 0.3,
      "lo": 0.10779126740630099,
      "hi": 0.6032218525388546,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "dateEra=old",
      "target": "diy:gemini-3.8-flash",
      "value": 0.26190476190476164,
      "lo": 0.1850234177739939,
      "hi": 0.35674800609347257,
      "n": 98,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "dateEra=old",
      "target": "diy:gemini-3.8-flash",
      "value": 0.24829931972789096,
      "lo": 0.1733514568660709,
      "hi": 0.34223547693667944,
      "n": 98,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "line_items_f1",
      "stratum": "dateEra=old",
      "target": "diy:gemini-3.8-flash",
      "value": 0.43902439024390244,
      "lo": 0.34283834586466166,
      "hi": 0.5,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "dateEra=old",
      "target": "diy:gemini-3.8-flash",
      "value": 0.26190476190476164,
      "lo": 0.1850234177739939,
      "hi": 0.35674800609347257,
      "n": 98,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "dateEra=recent",
      "target": "diy:gemini-3.8-flash",
      "value": 0.28205128205128205,
      "lo": 0.11117351106536214,
      "hi": 0.5523552409864679,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "dateEra=recent",
      "target": "diy:gemini-3.8-flash",
      "value": 0.30434782608695643,
      "lo": 0.15604024453214013,
      "hi": 0.5086575626875919,
      "n": 23,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "dateEra=recent",
      "target": "diy:gemini-3.8-flash",
      "value": 0.3095238095238095,
      "lo": 0.13213685461323155,
      "hi": 0.5689339273820583,
      "n": 14,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.tax",
      "stratum": "dateEra=recent",
      "target": "diy:gemini-3.8-flash",
      "value": 0.26666666666666666,
      "lo": 0.08962557199115637,
      "hi": 0.5732230679448586,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "dateEra=recent",
      "target": "diy:gemini-3.8-flash",
      "value": 0.30434782608695643,
      "lo": 0.15604024453214013,
      "hi": 0.5086575626875919,
      "n": 23,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "dateEra=recent",
      "target": "diy:gemini-3.8-flash",
      "value": 0.30434782608695643,
      "lo": 0.15604024453214013,
      "hi": 0.5086575626875919,
      "n": 23,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "line_items_f1",
      "stratum": "dateEra=recent",
      "target": "diy:gemini-3.8-flash",
      "value": 0.36904761904761907,
      "lo": 0.21714285714285714,
      "hi": 0.5,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "dateEra=recent",
      "target": "diy:gemini-3.8-flash",
      "value": 0.30434782608695643,
      "lo": 0.15604024453214013,
      "hi": 0.5086575626875919,
      "n": 23,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "docType=receipt",
      "target": "diy:gemini-3.8-flash",
      "value": 0.25964912280701735,
      "lo": 0.18205180924999986,
      "hi": 0.35592883964441313,
      "n": 95,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "docType=receipt",
      "target": "diy:gemini-3.8-flash",
      "value": 0.2619047619047616,
      "lo": 0.1893467213237418,
      "hi": 0.35025392013670703,
      "n": 112,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "docType=receipt",
      "target": "diy:gemini-3.8-flash",
      "value": 0.26515151515151497,
      "lo": 0.18416519784516566,
      "hi": 0.36578387892167735,
      "n": 88,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.subtotal",
      "stratum": "docType=receipt",
      "target": "diy:gemini-3.8-flash",
      "value": 0.22916666666666669,
      "lo": 0.08937790137618432,
      "hi": 0.4738262589133022,
      "n": 16,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.tax",
      "stratum": "docType=receipt",
      "target": "diy:gemini-3.8-flash",
      "value": 0.28333333333333327,
      "lo": 0.1340262550429334,
      "hi": 0.5024613134466216,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "docType=receipt",
      "target": "diy:gemini-3.8-flash",
      "value": 0.2699724517906332,
      "lo": 0.19885456121008174,
      "hi": 0.3552465587822836,
      "n": 121,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.receipt_number",
      "stratum": "docType=receipt",
      "target": "diy:gemini-3.8-flash",
      "value": 0.2727272727272727,
      "lo": 0.09746059290024889,
      "hi": 0.5656453011761291,
      "n": 11,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "docType=receipt",
      "target": "diy:gemini-3.8-flash",
      "value": 0.25895316804407675,
      "lo": 0.18917059682912157,
      "hi": 0.3435700977736565,
      "n": 121,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "line_items_f1",
      "stratum": "docType=receipt",
      "target": "diy:gemini-3.8-flash",
      "value": 0.4036144578313253,
      "lo": 0.3141025641025641,
      "hi": 0.47619047619047616,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "docType=receipt",
      "target": "diy:gemini-3.8-flash",
      "value": 0.2699724517906332,
      "lo": 0.19885456121008174,
      "hi": 0.3552465587822836,
      "n": 121,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "lang=en",
      "target": "diy:gemini-3.8-flash",
      "value": 0.2570281124497991,
      "lo": 0.17524011507751108,
      "hi": 0.36031198318224317,
      "n": 83,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "lang=en",
      "target": "diy:gemini-3.8-flash",
      "value": 0.2518518518518517,
      "lo": 0.17359965938733746,
      "hi": 0.3504202436705133,
      "n": 90,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "lang=en",
      "target": "diy:gemini-3.8-flash",
      "value": 0.25675675675675674,
      "lo": 0.17097707299565887,
      "hi": 0.36654444130528563,
      "n": 74,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.tax",
      "stratum": "lang=en",
      "target": "diy:gemini-3.8-flash",
      "value": 0.3,
      "lo": 0.10779126740630099,
      "hi": 0.6032218525388546,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "lang=en",
      "target": "diy:gemini-3.8-flash",
      "value": 0.2626262626262624,
      "lo": 0.18598149683086684,
      "hi": 0.357004371524566,
      "n": 99,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "lang=en",
      "target": "diy:gemini-3.8-flash",
      "value": 0.24915824915824897,
      "lo": 0.1744103868950867,
      "hi": 0.34264560165072544,
      "n": 99,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "line_items_f1",
      "stratum": "lang=en",
      "target": "diy:gemini-3.8-flash",
      "value": 0.4375,
      "lo": 0.33739583333333356,
      "hi": 0.5,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "lang=en",
      "target": "diy:gemini-3.8-flash",
      "value": 0.2626262626262624,
      "lo": 0.18598149683086684,
      "hi": 0.357004371524566,
      "n": 99,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "provenance=intact",
      "target": "diy:gemini-3.8-flash",
      "value": 0.19607843137254902,
      "lo": 0.07267421908451976,
      "hi": 0.43151915709733135,
      "n": 17,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "provenance=intact",
      "target": "diy:gemini-3.8-flash",
      "value": 0.16666666666666669,
      "lo": 0.06143139945263601,
      "hi": 0.3793187059159103,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "provenance=intact",
      "target": "diy:gemini-3.8-flash",
      "value": 0.13888888888888887,
      "lo": 0.03496201151166098,
      "hi": 0.41795032662823517,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.tax",
      "stratum": "provenance=intact",
      "target": "diy:gemini-3.8-flash",
      "value": 0.33333333333333337,
      "lo": 0.1269754121301785,
      "hi": 0.632202187824118,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "provenance=intact",
      "target": "diy:gemini-3.8-flash",
      "value": 0.16666666666666669,
      "lo": 0.06143139945263601,
      "hi": 0.3793187059159103,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "provenance=intact",
      "target": "diy:gemini-3.8-flash",
      "value": 0.15,
      "lo": 0.052368745896216595,
      "hi": 0.36041886474075696,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "line_items_f1",
      "stratum": "provenance=intact",
      "target": "diy:gemini-3.8-flash",
      "value": 0.47674418604651164,
      "lo": 0.4268292682926829,
      "hi": 0.5,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "provenance=intact",
      "target": "diy:gemini-3.8-flash",
      "value": 0.16666666666666669,
      "lo": 0.06143139945263601,
      "hi": 0.3793187059159103,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "provenance=none",
      "target": "diy:gemini-3.8-flash",
      "value": 0.327683615819209,
      "lo": 0.22169172275612037,
      "hi": 0.4547426930751457,
      "n": 59,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "provenance=none",
      "target": "diy:gemini-3.8-flash",
      "value": 0.3287037037037035,
      "lo": 0.23130535101308564,
      "hi": 0.4434547729319967,
      "n": 72,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "provenance=none",
      "target": "diy:gemini-3.8-flash",
      "value": 0.33333333333333326,
      "lo": 0.22729261523522676,
      "hi": 0.45943133547880305,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "provenance=none",
      "target": "diy:gemini-3.8-flash",
      "value": 0.333333333333333,
      "lo": 0.2402877124166734,
      "hi": 0.4414716493941309,
      "n": 81,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "provenance=none",
      "target": "diy:gemini-3.8-flash",
      "value": 0.3251028806584359,
      "lo": 0.2330357855180404,
      "hi": 0.4330079892216927,
      "n": 81,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "provenance=none",
      "target": "diy:gemini-3.8-flash",
      "value": 0.333333333333333,
      "lo": 0.2402877124166734,
      "hi": 0.4414716493941309,
      "n": 81,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "provenance=stripped",
      "target": "diy:gemini-3.8-flash",
      "value": 0.10526315789473684,
      "lo": 0.02935856029887013,
      "hi": 0.31394082711109006,
      "n": 19,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "provenance=stripped",
      "target": "diy:gemini-3.8-flash",
      "value": 0.11666666666666665,
      "lo": 0.035533746172033476,
      "hi": 0.3213288750017947,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "provenance=stripped",
      "target": "diy:gemini-3.8-flash",
      "value": 0.10416666666666666,
      "lo": 0.02607804969976618,
      "hi": 0.3355280307233294,
      "n": 16,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.tax",
      "stratum": "provenance=stripped",
      "target": "diy:gemini-3.8-flash",
      "value": 0.2333333333333333,
      "lo": 0.07255530696407597,
      "hi": 0.5421288529627982,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "provenance=stripped",
      "target": "diy:gemini-3.8-flash",
      "value": 0.11666666666666665,
      "lo": 0.035533746172033476,
      "hi": 0.3213288750017947,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "provenance=stripped",
      "target": "diy:gemini-3.8-flash",
      "value": 0.09999999999999999,
      "lo": 0.027866481213768224,
      "hi": 0.3010336452284873,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "line_items_f1",
      "stratum": "provenance=stripped",
      "target": "diy:gemini-3.8-flash",
      "value": 0.325,
      "lo": 0.17100166790214982,
      "hi": 0.4581005586592179,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "provenance=stripped",
      "target": "diy:gemini-3.8-flash",
      "value": 0.11666666666666665,
      "lo": 0.035533746172033476,
      "hi": 0.3213288750017947,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "source=commons",
      "target": "diy:gemini-3.8-flash",
      "value": 0.3333333333333333,
      "lo": 0.1430225574285459,
      "hi": 0.5996759000228535,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "source=commons",
      "target": "diy:gemini-3.8-flash",
      "value": 0.3333333333333333,
      "lo": 0.1430225574285459,
      "hi": 0.5996759000228535,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "source=commons",
      "target": "diy:gemini-3.8-flash",
      "value": 0.3333333333333333,
      "lo": 0.1430225574285459,
      "hi": 0.5996759000228535,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "source=commons",
      "target": "diy:gemini-3.8-flash",
      "value": 0.3333333333333333,
      "lo": 0.1430225574285459,
      "hi": 0.5996759000228535,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "source=edited:sharp-patch",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.2775327998628892,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "source=edited:sharp-patch",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.2775327998628892,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "source=edited:sharp-patch",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.2775327998628892,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "source=edited:sharp-patch",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.2775327998628892,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.merchant",
      "stratum": "source=sroie",
      "target": "diy:gemini-3.8-flash",
      "value": 0.3272727272727273,
      "lo": 0.21813065595168943,
      "hi": 0.45896776443421317,
      "n": 55,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.date",
      "stratum": "source=sroie",
      "target": "diy:gemini-3.8-flash",
      "value": 0.327683615819209,
      "lo": 0.22169172275612037,
      "hi": 0.4547426930751457,
      "n": 59,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.currency",
      "stratum": "source=sroie",
      "target": "diy:gemini-3.8-flash",
      "value": 0.3333333333333334,
      "lo": 0.22062697673438705,
      "hi": 0.46897043530690063,
      "n": 52,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "field_accuracy.total",
      "stratum": "source=sroie",
      "target": "diy:gemini-3.8-flash",
      "value": 0.33333333333333315,
      "lo": 0.23287456398365092,
      "hi": 0.451615881523039,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "all_key_fields",
      "stratum": "source=sroie",
      "target": "diy:gemini-3.8-flash",
      "value": 0.32352941176470573,
      "lo": 0.2243760409532998,
      "hi": 0.4415550189949602,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "extraction",
      "metric": "valid_json",
      "stratum": "source=sroie",
      "target": "diy:gemini-3.8-flash",
      "value": 0.33333333333333315,
      "lo": 0.23287456398365092,
      "hi": 0.451615881523039,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "all",
      "target": "diy:gemini-3.8-flash",
      "value": 0.9554012345679013,
      "lo": 0.8982986111111111,
      "hi": 1,
      "n": 121,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "all",
      "target": "diy:gemini-3.8-flash",
      "value": 0.2,
      "lo": 0.1,
      "hi": 1,
      "n": 40,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "all",
      "target": "diy:gemini-3.8-flash",
      "value": 0.95,
      "lo": 0.875,
      "hi": 1,
      "n": 40,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "all",
      "target": "diy:gemini-3.8-flash",
      "value": 0.95,
      "lo": 0.875,
      "hi": 1,
      "n": 40,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "all",
      "target": "diy:gemini-3.8-flash",
      "value": 0.12499999999999997,
      "lo": 0.054595002509453996,
      "hi": 0.26112119838851094,
      "n": 40,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_5",
      "stratum": "all",
      "target": "diy:gemini-3.8-flash",
      "value": 0.004115226337448559,
      "lo": 0.0003060448651870523,
      "hi": 0.052829834102763824,
      "n": 81,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "all",
      "target": "diy:gemini-3.8-flash",
      "value": 0.11666666666666665,
      "lo": 0.04946771889583815,
      "hi": 0.25104217535541495,
      "n": 40,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_7",
      "stratum": "all",
      "target": "diy:gemini-3.8-flash",
      "value": 0.004115226337448559,
      "lo": 0.0003060448651870523,
      "hi": 0.052829834102763824,
      "n": 81,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "base=sroie",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.16112515805281938,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "base=sroie",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.16112515805281938,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "capture=photo",
      "target": "diy:gemini-3.8-flash",
      "value": 1,
      "lo": 1,
      "hi": 1,
      "n": 21,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "capture=photo",
      "target": "diy:gemini-3.8-flash",
      "value": 1,
      "lo": 1,
      "hi": 1,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "capture=photo",
      "target": "diy:gemini-3.8-flash",
      "value": 1,
      "lo": 1,
      "hi": 1,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "capture=photo",
      "target": "diy:gemini-3.8-flash",
      "value": 1,
      "lo": 1,
      "hi": 1,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "capture=photo",
      "target": "diy:gemini-3.8-flash",
      "value": 0.2777777777777778,
      "lo": 0.10461389175889282,
      "hi": 0.5587167778656588,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "capture=photo",
      "target": "diy:gemini-3.8-flash",
      "value": 0.25,
      "lo": 0.08894166839405471,
      "hi": 0.5323053349335657,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "capture=scan",
      "target": "diy:gemini-3.8-flash",
      "value": 0.9486921529175051,
      "lo": 0.8631790744466801,
      "hi": 1,
      "n": 99,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "capture=scan",
      "target": "diy:gemini-3.8-flash",
      "value": 0.9285714285714286,
      "lo": 0.8214285714285714,
      "hi": 1,
      "n": 28,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "capture=scan",
      "target": "diy:gemini-3.8-flash",
      "value": 0.9285714285714286,
      "lo": 0.8214285714285714,
      "hi": 1,
      "n": 28,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "capture=scan",
      "target": "diy:gemini-3.8-flash",
      "value": 0.9285714285714286,
      "lo": 0.8214285714285714,
      "hi": 1,
      "n": 28,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "capture=scan",
      "target": "diy:gemini-3.8-flash",
      "value": 0.05952380952380952,
      "lo": 0.014799033820906068,
      "hi": 0.2105295918061484,
      "n": 28,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_5",
      "stratum": "capture=scan",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.05132795219689034,
      "n": 71,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "capture=scan",
      "target": "diy:gemini-3.8-flash",
      "value": 0.05952380952380952,
      "lo": 0.014799033820906068,
      "hi": 0.2105295918061484,
      "n": 28,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_7",
      "stratum": "capture=scan",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.05132795219689034,
      "n": 71,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "country=MY",
      "target": "diy:gemini-3.8-flash",
      "value": 0.9426470588235294,
      "lo": 0.8494117647058823,
      "hi": 1,
      "n": 93,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "country=MY",
      "target": "diy:gemini-3.8-flash",
      "value": 0.92,
      "lo": 0.8,
      "hi": 1,
      "n": 25,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "country=MY",
      "target": "diy:gemini-3.8-flash",
      "value": 0.92,
      "lo": 0.8,
      "hi": 1,
      "n": 25,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "country=MY",
      "target": "diy:gemini-3.8-flash",
      "value": 0.92,
      "lo": 0.8,
      "hi": 1,
      "n": 25,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "country=MY",
      "target": "diy:gemini-3.8-flash",
      "value": 0.026666666666666665,
      "lo": 0.003503883757309423,
      "hi": 0.17591811379832314,
      "n": 25,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_5",
      "stratum": "country=MY",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.053471336520070537,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "country=MY",
      "target": "diy:gemini-3.8-flash",
      "value": 0.026666666666666665,
      "lo": 0.003503883757309423,
      "hi": 0.17591811379832314,
      "n": 25,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_7",
      "stratum": "country=MY",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.053471336520070537,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "currency=MYR",
      "target": "diy:gemini-3.8-flash",
      "value": 0.9426470588235294,
      "lo": 0.851764705882353,
      "hi": 1,
      "n": 93,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "currency=MYR",
      "target": "diy:gemini-3.8-flash",
      "value": 0.92,
      "lo": 0.8,
      "hi": 1,
      "n": 25,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "currency=MYR",
      "target": "diy:gemini-3.8-flash",
      "value": 0.92,
      "lo": 0.8,
      "hi": 1,
      "n": 25,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "currency=MYR",
      "target": "diy:gemini-3.8-flash",
      "value": 0.92,
      "lo": 0.8,
      "hi": 1,
      "n": 25,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "currency=MYR",
      "target": "diy:gemini-3.8-flash",
      "value": 0.026666666666666665,
      "lo": 0.003503883757309423,
      "hi": 0.17591811379832314,
      "n": 25,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_5",
      "stratum": "currency=MYR",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.053471336520070537,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "currency=MYR",
      "target": "diy:gemini-3.8-flash",
      "value": 0.026666666666666665,
      "lo": 0.003503883757309423,
      "hi": 0.17591811379832314,
      "n": 25,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_7",
      "stratum": "currency=MYR",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.053471336520070537,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "dateEra=old",
      "target": "diy:gemini-3.8-flash",
      "value": 0.9522058823529411,
      "lo": 0.8725490196078431,
      "hi": 1,
      "n": 98,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "dateEra=old",
      "target": "diy:gemini-3.8-flash",
      "value": 0.9333333333333333,
      "lo": 0.8333333333333334,
      "hi": 1,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "dateEra=old",
      "target": "diy:gemini-3.8-flash",
      "value": 0.9333333333333333,
      "lo": 0.8333333333333334,
      "hi": 1,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "dateEra=old",
      "target": "diy:gemini-3.8-flash",
      "value": 0.9333333333333333,
      "lo": 0.8333333333333334,
      "hi": 1,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "dateEra=old",
      "target": "diy:gemini-3.8-flash",
      "value": 0.07777777777777777,
      "lo": 0.023533178585232586,
      "hi": 0.22787813120611877,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_5",
      "stratum": "dateEra=old",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.053471336520070537,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "dateEra=old",
      "target": "diy:gemini-3.8-flash",
      "value": 0.06666666666666667,
      "lo": 0.018477023791270378,
      "hi": 0.2132345836261692,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_7",
      "stratum": "dateEra=old",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.053471336520070537,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "dateEra=recent",
      "target": "diy:gemini-3.8-flash",
      "value": 0.9769230769230769,
      "lo": 0.9076923076923077,
      "hi": 1,
      "n": 23,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "dateEra=recent",
      "target": "diy:gemini-3.8-flash",
      "value": 0.7,
      "lo": 0.5,
      "hi": 1,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "dateEra=recent",
      "target": "diy:gemini-3.8-flash",
      "value": 0.7,
      "lo": 0.5,
      "hi": 1,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "dateEra=recent",
      "target": "diy:gemini-3.8-flash",
      "value": 1,
      "lo": 0.5,
      "hi": 1,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "dateEra=recent",
      "target": "diy:gemini-3.8-flash",
      "value": 0.26666666666666666,
      "lo": 0.08962557199115637,
      "hi": 0.5732230679448586,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_5",
      "stratum": "dateEra=recent",
      "target": "diy:gemini-3.8-flash",
      "value": 0.02564102564102564,
      "lo": 0.001909533775505512,
      "hi": 0.2657706912784776,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "dateEra=recent",
      "target": "diy:gemini-3.8-flash",
      "value": 0.26666666666666666,
      "lo": 0.08962557199115637,
      "hi": 0.5732230679448586,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_7",
      "stratum": "dateEra=recent",
      "target": "diy:gemini-3.8-flash",
      "value": 0.02564102564102564,
      "lo": 0.001909533775505512,
      "hi": 0.2657706912784776,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "docType=receipt",
      "target": "diy:gemini-3.8-flash",
      "value": 0.9554012345679013,
      "lo": 0.8942669753086421,
      "hi": 1,
      "n": 121,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "docType=receipt",
      "target": "diy:gemini-3.8-flash",
      "value": 0.2,
      "lo": 0.1,
      "hi": 1,
      "n": 40,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "docType=receipt",
      "target": "diy:gemini-3.8-flash",
      "value": 0.95,
      "lo": 0.875,
      "hi": 1,
      "n": 40,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "docType=receipt",
      "target": "diy:gemini-3.8-flash",
      "value": 0.95,
      "lo": 0.875,
      "hi": 1,
      "n": 40,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "docType=receipt",
      "target": "diy:gemini-3.8-flash",
      "value": 0.12499999999999997,
      "lo": 0.054595002509453996,
      "hi": 0.26112119838851094,
      "n": 40,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_5",
      "stratum": "docType=receipt",
      "target": "diy:gemini-3.8-flash",
      "value": 0.004115226337448559,
      "lo": 0.0003060448651870523,
      "hi": 0.052829834102763824,
      "n": 81,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "docType=receipt",
      "target": "diy:gemini-3.8-flash",
      "value": 0.11666666666666665,
      "lo": 0.04946771889583815,
      "hi": 0.25104217535541495,
      "n": 40,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_7",
      "stratum": "docType=receipt",
      "target": "diy:gemini-3.8-flash",
      "value": 0.004115226337448559,
      "lo": 0.0003060448651870523,
      "hi": 0.052829834102763824,
      "n": 81,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "lang=en",
      "target": "diy:gemini-3.8-flash",
      "value": 0.9521739130434783,
      "lo": 0.8705314009661835,
      "hi": 1,
      "n": 99,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "lang=en",
      "target": "diy:gemini-3.8-flash",
      "value": 0.9333333333333333,
      "lo": 0.8333333333333334,
      "hi": 1,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "lang=en",
      "target": "diy:gemini-3.8-flash",
      "value": 0.9333333333333333,
      "lo": 0.8333333333333334,
      "hi": 1,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "lang=en",
      "target": "diy:gemini-3.8-flash",
      "value": 0.9333333333333333,
      "lo": 0.8333333333333334,
      "hi": 1,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "lang=en",
      "target": "diy:gemini-3.8-flash",
      "value": 0.07777777777777777,
      "lo": 0.023533178585232586,
      "hi": 0.22787813120611877,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_5",
      "stratum": "lang=en",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 3.469446951953614e-18,
      "hi": 0.05273725818905173,
      "n": 69,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "lang=en",
      "target": "diy:gemini-3.8-flash",
      "value": 0.06666666666666667,
      "lo": 0.018477023791270378,
      "hi": 0.2132345836261692,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_7",
      "stratum": "lang=en",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 3.469446951953614e-18,
      "hi": 0.05273725818905173,
      "n": 69,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_5",
      "stratum": "source=commons",
      "target": "diy:gemini-3.8-flash",
      "value": 0.02564102564102564,
      "lo": 0.001909533775505512,
      "hi": 0.2657706912784776,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_7",
      "stratum": "source=commons",
      "target": "diy:gemini-3.8-flash",
      "value": 0.02564102564102564,
      "lo": 0.001909533775505512,
      "hi": 0.2657706912784776,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "source=edited:sharp-patch",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.2775327998628892,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "source=edited:sharp-patch",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.2775327998628892,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_5",
      "stratum": "source=sroie",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.053471336520070537,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_7",
      "stratum": "source=sroie",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.053471336520070537,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "provenance=intact",
      "target": "diy:gemini-3.8-flash",
      "value": 0.15,
      "lo": 0.052368745896216595,
      "hi": 0.36041886474075696,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "provenance=intact",
      "target": "diy:gemini-3.8-flash",
      "value": 0.13333333333333333,
      "lo": 0.043720722486203234,
      "hi": 0.3411043934191976,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "provenance=stripped",
      "target": "diy:gemini-3.8-flash",
      "value": 0.09999999999999999,
      "lo": 0.027866481213768224,
      "hi": 0.3010336452284873,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "provenance=stripped",
      "target": "diy:gemini-3.8-flash",
      "value": 0.09999999999999999,
      "lo": 0.027866481213768224,
      "hi": 0.3010336452284873,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_5",
      "stratum": "provenance=none",
      "target": "diy:gemini-3.8-flash",
      "value": 0.004115226337448559,
      "lo": 0.0003060448651870523,
      "hi": 0.052829834102763824,
      "n": 81,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_7",
      "stratum": "provenance=none",
      "target": "diy:gemini-3.8-flash",
      "value": 0.004115226337448559,
      "lo": 0.0003060448651870523,
      "hi": 0.052829834102763824,
      "n": 81,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "class=ai_full",
      "target": "diy:gemini-3.8-flash",
      "value": 0.9231481481481482,
      "lo": 0.812037037037037,
      "hi": 1,
      "n": 101,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "class=ai_full",
      "target": "diy:gemini-3.8-flash",
      "value": 0.4,
      "lo": 0.2,
      "hi": 1,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "class=ai_full",
      "target": "diy:gemini-3.8-flash",
      "value": 0.9,
      "lo": 0.75,
      "hi": 1,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "class=ai_full",
      "target": "diy:gemini-3.8-flash",
      "value": 0.9,
      "lo": 0.75,
      "hi": 1,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "class=ai_full",
      "target": "diy:gemini-3.8-flash",
      "value": 0.24999999999999994,
      "lo": 0.11186170140766563,
      "hi": 0.468700877618744,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "class=ai_full",
      "target": "diy:gemini-3.8-flash",
      "value": 0.2333333333333333,
      "lo": 0.1011725498384442,
      "hi": 0.4514275344563928,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "class=ai_inpaint",
      "target": "diy:gemini-3.8-flash",
      "value": 0.9876543209876543,
      "lo": 0.9629629629629629,
      "hi": 1,
      "n": 91,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "class=ai_inpaint",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 1,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "class=ai_inpaint",
      "target": "diy:gemini-3.8-flash",
      "value": 1,
      "lo": 1,
      "hi": 1,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "class=ai_inpaint",
      "target": "diy:gemini-3.8-flash",
      "value": 1,
      "lo": 1,
      "hi": 1,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "class=ai_inpaint",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.2775327998628892,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "class=ai_inpaint",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.2775327998628892,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "class=manual_edit",
      "target": "diy:gemini-3.8-flash",
      "value": 0.9876543209876543,
      "lo": 0.9629629629629629,
      "hi": 1,
      "n": 91,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "class=manual_edit",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 1,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "class=manual_edit",
      "target": "diy:gemini-3.8-flash",
      "value": 1,
      "lo": 1,
      "hi": 1,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "class=manual_edit",
      "target": "diy:gemini-3.8-flash",
      "value": 1,
      "lo": 1,
      "hi": 1,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "class=manual_edit",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.2775327998628892,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "class=manual_edit",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.2775327998628892,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "arithmetic=consistent",
      "target": "diy:gemini-3.8-flash",
      "value": 0.9267676767676768,
      "lo": 0.819577721661055,
      "hi": 1,
      "n": 103,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "arithmetic=consistent",
      "target": "diy:gemini-3.8-flash",
      "value": 0.18181818181818182,
      "lo": 0.045454545454545456,
      "hi": 1,
      "n": 22,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "arithmetic=consistent",
      "target": "diy:gemini-3.8-flash",
      "value": 0.9090909090909091,
      "lo": 0.7727272727272727,
      "hi": 1,
      "n": 22,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "arithmetic=consistent",
      "target": "diy:gemini-3.8-flash",
      "value": 0.9090909090909091,
      "lo": 0.7727272727272727,
      "hi": 1,
      "n": 22,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "arithmetic=consistent",
      "target": "diy:gemini-3.8-flash",
      "value": 0.15151515151515152,
      "lo": 0.05568961722175389,
      "hi": 0.3509486287175879,
      "n": 22,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "arithmetic=consistent",
      "target": "diy:gemini-3.8-flash",
      "value": 0.13636363636363635,
      "lo": 0.04749003714055253,
      "hi": 0.33334987166571717,
      "n": 22,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "arithmetic=inconsistent",
      "target": "diy:gemini-3.8-flash",
      "value": 0.99039780521262,
      "lo": 0.9690843621399178,
      "hi": 1,
      "n": 99,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "arithmetic=inconsistent",
      "target": "diy:gemini-3.8-flash",
      "value": 0.2222222222222222,
      "lo": 0.05555555555555555,
      "hi": 1,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "arithmetic=inconsistent",
      "target": "diy:gemini-3.8-flash",
      "value": 1,
      "lo": 1,
      "hi": 1,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "arithmetic=inconsistent",
      "target": "diy:gemini-3.8-flash",
      "value": 1,
      "lo": 1,
      "hi": 1,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "arithmetic=inconsistent",
      "target": "diy:gemini-3.8-flash",
      "value": 0.09259259259259259,
      "lo": 0.02313862362313196,
      "hi": 0.305355558607478,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "arithmetic=inconsistent",
      "target": "diy:gemini-3.8-flash",
      "value": 0.09259259259259259,
      "lo": 0.02313862362313196,
      "hi": 0.305355558607478,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "decisions",
      "metric": "decision_consistency",
      "stratum": "all",
      "target": "diy:gemini-3.8-flash",
      "value": 0.8677685950413223,
      "lo": 0.7959633756720691,
      "hi": 0.9169408217054055,
      "n": 121,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "decisions",
      "metric": "total_consistency",
      "stratum": "all",
      "target": "diy:gemini-3.8-flash",
      "value": 0.19008264462809918,
      "lo": 0.1301342764638027,
      "hi": 0.2691037594540558,
      "n": 121,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "decisions",
      "metric": "calibration_error",
      "stratum": "all",
      "target": "diy:gemini-3.8-flash",
      "value": 0.03530612244897954,
      "lo": 0.01698977640482497,
      "hi": 0.07141895450470633,
      "n": 121,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_5",
      "stratum": "all",
      "target": "diy:gemini-3.8-flash",
      "value": 944.2592592592594,
      "lo": 841.6666666666667,
      "hi": 1107.777777777778,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_7",
      "stratum": "all",
      "target": "diy:gemini-3.8-flash",
      "value": 952.5925925925926,
      "lo": 850.0000000000001,
      "hi": 1124.4444444444443,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_5",
      "stratum": "assumption=fakes_0_1pct",
      "target": "diy:gemini-3.8-flash",
      "value": 157.38888888888889,
      "lo": 84.14583333333339,
      "hi": 299.6666666666667,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_7",
      "stratum": "assumption=fakes_0_1pct",
      "target": "diy:gemini-3.8-flash",
      "value": 158.2222222222222,
      "lo": 84.9791666666667,
      "hi": 301.3333333333333,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_5",
      "stratum": "assumption=fakes_5pct",
      "target": "diy:gemini-3.8-flash",
      "value": 4441.460905349794,
      "lo": 4149.794238683128,
      "hi": 4733.127572016461,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_7",
      "stratum": "assumption=fakes_5pct",
      "target": "diy:gemini-3.8-flash",
      "value": 4483.127572016461,
      "lo": 4208.333333333334,
      "hi": 4757.9218106995895,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_5",
      "stratum": "assumption=fake_32usd",
      "target": "diy:gemini-3.8-flash",
      "value": 349.2592592592593,
      "lo": 266.6666666666667,
      "hi": 495.77777777777777,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_7",
      "stratum": "assumption=fake_32usd",
      "target": "diy:gemini-3.8-flash",
      "value": 351.9259259259259,
      "lo": 272,
      "hi": 498.44444444444446,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_5",
      "stratum": "assumption=fake_182usd",
      "target": "diy:gemini-3.8-flash",
      "value": 1661.7592592592596,
      "lo": 1516.6666666666665,
      "hi": 1845.777777777778,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_7",
      "stratum": "assumption=fake_182usd",
      "target": "diy:gemini-3.8-flash",
      "value": 1676.925925925926,
      "lo": 1531.8333333333337,
      "hi": 1860.9444444444446,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_5",
      "stratum": "assumption=review_8usd",
      "target": "diy:gemini-3.8-flash",
      "value": 907.5925925925927,
      "lo": 833.3333333333334,
      "hi": 997.0555555555555,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_7",
      "stratum": "assumption=review_8usd",
      "target": "diy:gemini-3.8-flash",
      "value": 915.925925925926,
      "lo": 841.6666666666667,
      "hi": 1006.1111111111112,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_5",
      "stratum": "assumption=review_73usd",
      "target": "diy:gemini-3.8-flash",
      "value": 1172.4074074074076,
      "lo": 833.3333333333334,
      "hi": 1800.5555555555554,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_7",
      "stratum": "assumption=review_73usd",
      "target": "diy:gemini-3.8-flash",
      "value": 1180.7407407407409,
      "lo": 850.0000000000001,
      "hi": 1800.5555555555554,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "approve_all_cost_per_1000",
      "stratum": "all",
      "target": "diy:gemini-3.8-flash",
      "value": 1000,
      "lo": 1000,
      "hi": 1000,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "approve_all_cost_per_1000",
      "stratum": "assumption=fakes_0_1pct",
      "target": "diy:gemini-3.8-flash",
      "value": 100,
      "lo": 100,
      "hi": 100,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "approve_all_cost_per_1000",
      "stratum": "assumption=fakes_5pct",
      "target": "diy:gemini-3.8-flash",
      "value": 5000,
      "lo": 5000,
      "hi": 5000,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "approve_all_cost_per_1000",
      "stratum": "assumption=fake_32usd",
      "target": "diy:gemini-3.8-flash",
      "value": 320,
      "lo": 320,
      "hi": 320,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "approve_all_cost_per_1000",
      "stratum": "assumption=fake_182usd",
      "target": "diy:gemini-3.8-flash",
      "value": 1820,
      "lo": 1820,
      "hi": 1820,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "approve_all_cost_per_1000",
      "stratum": "assumption=review_8usd",
      "target": "diy:gemini-3.8-flash",
      "value": 1000,
      "lo": 1000,
      "hi": 1000,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "approve_all_cost_per_1000",
      "stratum": "assumption=review_73usd",
      "target": "diy:gemini-3.8-flash",
      "value": 1000,
      "lo": 1000,
      "hi": 1000,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "evidence",
      "metric": "edited_field_named",
      "stratum": "all",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.16112515805281938,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "all",
      "target": "diy:gemini-3.8-flash",
      "value": 6522.5,
      "lo": 6522.5,
      "hi": 6522.5,
      "n": 98,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "all",
      "target": "diy:gemini-3.8-flash",
      "value": 16140.199999999986,
      "lo": 16140.199999999986,
      "hi": 16140.199999999986,
      "n": 98,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "all",
      "target": "diy:gemini-3.8-flash",
      "value": 26382.780000000006,
      "lo": 26382.780000000006,
      "hi": 26382.780000000006,
      "n": 98,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "all",
      "target": "diy:gemini-3.8-flash",
      "value": 0.004439181122448979,
      "lo": 0.004439181122448979,
      "hi": 0.004439181122448979,
      "n": 98,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "all",
      "target": "diy:gemini-3.8-flash",
      "value": 0.004439181122448979,
      "lo": 0.004439181122448979,
      "hi": 0.004439181122448979,
      "n": 98,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "all",
      "target": "diy:gemini-3.8-flash",
      "value": 0.7836644591611479,
      "lo": 0.7434472034386128,
      "hi": 0.8191111962815625,
      "n": 453,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "all",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.008408735123582253,
      "n": 453,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "all",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.008408735123582253,
      "n": 453,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "arithmetic=consistent",
      "target": "diy:gemini-3.8-flash",
      "value": 6421,
      "lo": 6421,
      "hi": 6421,
      "n": 90,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "arithmetic=consistent",
      "target": "diy:gemini-3.8-flash",
      "value": 12071.599999999999,
      "lo": 12071.599999999999,
      "hi": 12071.599999999999,
      "n": 90,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "arithmetic=consistent",
      "target": "diy:gemini-3.8-flash",
      "value": 22369.559999999994,
      "lo": 22369.559999999994,
      "hi": 22369.559999999994,
      "n": 90,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "arithmetic=consistent",
      "target": "diy:gemini-3.8-flash",
      "value": 0.004218291666666665,
      "lo": 0.004218291666666665,
      "hi": 0.004218291666666665,
      "n": 90,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "arithmetic=consistent",
      "target": "diy:gemini-3.8-flash",
      "value": 0.004218291666666665,
      "lo": 0.004218291666666665,
      "hi": 0.004218291666666665,
      "n": 90,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "arithmetic=consistent",
      "target": "diy:gemini-3.8-flash",
      "value": 0.7692307692307693,
      "lo": 0.7249112949346743,
      "hi": 0.8082981864358678,
      "n": 390,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "arithmetic=consistent",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.009753820311850517,
      "n": 390,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "arithmetic=consistent",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.009753820311850517,
      "n": 390,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "arithmetic=inconsistent",
      "target": "diy:gemini-3.8-flash",
      "value": 0.9074074074074074,
      "lo": 0.8009011740903732,
      "hi": 0.9597988652429706,
      "n": 54,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "arithmetic=inconsistent",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.0664135880908964,
      "n": 54,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "arithmetic=inconsistent",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.0664135880908964,
      "n": 54,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "arithmeticError=totals",
      "target": "diy:gemini-3.8-flash",
      "value": 0.7619047619047619,
      "lo": 0.5490882631704924,
      "hi": 0.8937198697736354,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "arithmeticError=totals",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.154639018924847,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "arithmeticError=totals",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.154639018924847,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "base=sroie",
      "target": "diy:gemini-3.8-flash",
      "value": 1,
      "lo": 0.93982814785791,
      "hi": 0.9999999999999998,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "base=sroie",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.06017185214208986,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "base=sroie",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.06017185214208986,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "capture=photo",
      "target": "diy:gemini-3.8-flash",
      "value": 10908,
      "lo": 10908,
      "hi": 10908,
      "n": 19,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "capture=photo",
      "target": "diy:gemini-3.8-flash",
      "value": 19424.599999999973,
      "lo": 19424.599999999973,
      "hi": 19424.599999999973,
      "n": 19,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "capture=photo",
      "target": "diy:gemini-3.8-flash",
      "value": 28968.920000000006,
      "lo": 28968.920000000006,
      "hi": 28968.920000000006,
      "n": 19,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "capture=photo",
      "target": "diy:gemini-3.8-flash",
      "value": 0.006375157894736843,
      "lo": 0.006375157894736843,
      "hi": 0.006375157894736843,
      "n": 19,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "capture=photo",
      "target": "diy:gemini-3.8-flash",
      "value": 0.006375157894736843,
      "lo": 0.006375157894736843,
      "hi": 0.006375157894736843,
      "n": 19,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "capture=photo",
      "target": "diy:gemini-3.8-flash",
      "value": 0.7466666666666667,
      "lo": 0.6378967313440042,
      "hi": 0.8313995063644785,
      "n": 75,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "capture=photo",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 3.469446951953614e-18,
      "hi": 0.04872384248280587,
      "n": 75,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "capture=photo",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 3.469446951953614e-18,
      "hi": 0.04872384248280587,
      "n": 75,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "capture=scan",
      "target": "diy:gemini-3.8-flash",
      "value": 5981.5,
      "lo": 5981.5,
      "hi": 5981.5,
      "n": 78,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "capture=scan",
      "target": "diy:gemini-3.8-flash",
      "value": 10110.249999999996,
      "lo": 10110.249999999996,
      "hi": 10110.249999999996,
      "n": 78,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "capture=scan",
      "target": "diy:gemini-3.8-flash",
      "value": 19565.420000000035,
      "lo": 19565.420000000035,
      "hi": 19565.420000000035,
      "n": 78,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "capture=scan",
      "target": "diy:gemini-3.8-flash",
      "value": 0.0038865576923076913,
      "lo": 0.0038865576923076913,
      "hi": 0.0038865576923076913,
      "n": 78,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "capture=scan",
      "target": "diy:gemini-3.8-flash",
      "value": 0.0038865576923076913,
      "lo": 0.0038865576923076913,
      "hi": 0.0038865576923076913,
      "n": 78,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "capture=scan",
      "target": "diy:gemini-3.8-flash",
      "value": 0.7833333333333333,
      "lo": 0.7379050408401375,
      "hi": 0.8227787269229361,
      "n": 360,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "capture=scan",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.010558056888693507,
      "n": 360,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "capture=scan",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.010558056888693507,
      "n": 360,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "capture=screenshot",
      "target": "diy:gemini-3.8-flash",
      "value": 1,
      "lo": 0.7961166989641515,
      "hi": 1,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "capture=screenshot",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "capture=screenshot",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "class=ai_full",
      "target": "diy:gemini-3.8-flash",
      "value": 11549,
      "lo": 11549,
      "hi": 11549,
      "n": 17,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "class=ai_full",
      "target": "diy:gemini-3.8-flash",
      "value": 27254.199999999997,
      "lo": 27254.199999999997,
      "hi": 27254.199999999997,
      "n": 17,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "class=ai_full",
      "target": "diy:gemini-3.8-flash",
      "value": 30534.84,
      "lo": 30534.84,
      "hi": 30534.84,
      "n": 17,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "class=ai_full",
      "target": "diy:gemini-3.8-flash",
      "value": 0.007206529411764705,
      "lo": 0.007206529411764705,
      "hi": 0.007206529411764705,
      "n": 17,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "class=ai_full",
      "target": "diy:gemini-3.8-flash",
      "value": 0.007206529411764705,
      "lo": 0.007206529411764705,
      "hi": 0.007206529411764705,
      "n": 17,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "class=ai_full",
      "target": "diy:gemini-3.8-flash",
      "value": 0.7166666666666667,
      "lo": 0.5923271483560882,
      "hi": 0.8149317157156729,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "class=ai_full",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.06017185214208986,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "class=ai_full",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.06017185214208986,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "class=ai_inpaint",
      "target": "diy:gemini-3.8-flash",
      "value": 1,
      "lo": 0.8864866068260312,
      "hi": 0.9999999999999999,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "class=ai_inpaint",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "class=ai_inpaint",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "class=duplicate",
      "target": "diy:gemini-3.8-flash",
      "value": 1,
      "lo": 0.9590643744356734,
      "hi": 1,
      "n": 90,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "class=duplicate",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.04093562556432678,
      "n": 90,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "class=duplicate",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.04093562556432678,
      "n": 90,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "class=genuine",
      "target": "diy:gemini-3.8-flash",
      "value": 6130,
      "lo": 6130,
      "hi": 6130,
      "n": 70,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "class=genuine",
      "target": "diy:gemini-3.8-flash",
      "value": 10745.999999999998,
      "lo": 10745.999999999998,
      "hi": 10745.999999999998,
      "n": 70,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "class=genuine",
      "target": "diy:gemini-3.8-flash",
      "value": 17398.45000000001,
      "lo": 17398.45000000001,
      "hi": 17398.45000000001,
      "n": 70,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "class=genuine",
      "target": "diy:gemini-3.8-flash",
      "value": 0.00390320357142857,
      "lo": 0.00390320357142857,
      "hi": 0.00390320357142857,
      "n": 70,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "class=genuine",
      "target": "diy:gemini-3.8-flash",
      "value": 0.00390320357142857,
      "lo": 0.00390320357142857,
      "hi": 0.00390320357142857,
      "n": 70,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "class=genuine",
      "target": "diy:gemini-3.8-flash",
      "value": 0.6666666666666666,
      "lo": 0.6004193945242017,
      "hi": 0.7269259218893706,
      "n": 210,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "class=genuine",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.017964050759283236,
      "n": 210,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "class=genuine",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.017964050759283236,
      "n": 210,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "class=hard_negative",
      "target": "diy:gemini-3.8-flash",
      "value": 5441,
      "lo": 5441,
      "hi": 5441,
      "n": 11,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "class=hard_negative",
      "target": "diy:gemini-3.8-flash",
      "value": 8215,
      "lo": 8215,
      "hi": 8215,
      "n": 11,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "class=hard_negative",
      "target": "diy:gemini-3.8-flash",
      "value": 9146.2,
      "lo": 9146.2,
      "hi": 9146.2,
      "n": 11,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "class=hard_negative",
      "target": "diy:gemini-3.8-flash",
      "value": 0.003573136363636363,
      "lo": 0.003573136363636363,
      "hi": 0.003573136363636363,
      "n": 11,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "class=hard_negative",
      "target": "diy:gemini-3.8-flash",
      "value": 0.003573136363636363,
      "lo": 0.003573136363636363,
      "hi": 0.003573136363636363,
      "n": 11,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "class=hard_negative",
      "target": "diy:gemini-3.8-flash",
      "value": 0.6666666666666666,
      "lo": 0.4960789442780561,
      "hi": 0.8024977232077931,
      "n": 33,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "class=hard_negative",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.10426999754245206,
      "n": 33,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "class=hard_negative",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.10426999754245206,
      "n": 33,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "class=manual_edit",
      "target": "diy:gemini-3.8-flash",
      "value": 1,
      "lo": 0.8864866068260312,
      "hi": 0.9999999999999999,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "class=manual_edit",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "class=manual_edit",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "country=JP",
      "target": "diy:gemini-3.8-flash",
      "value": 0.6666666666666666,
      "lo": 0.39062208887279953,
      "hi": 0.8618799089087867,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "country=JP",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "country=JP",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "country=MY",
      "target": "diy:gemini-3.8-flash",
      "value": 5981.5,
      "lo": 5981.5,
      "hi": 5981.5,
      "n": 72,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "country=MY",
      "target": "diy:gemini-3.8-flash",
      "value": 9944.100000000002,
      "lo": 9944.100000000002,
      "hi": 9944.100000000002,
      "n": 72,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "country=MY",
      "target": "diy:gemini-3.8-flash",
      "value": 20084.660000000054,
      "lo": 20084.660000000054,
      "hi": 20084.660000000054,
      "n": 72,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "country=MY",
      "target": "diy:gemini-3.8-flash",
      "value": 0.003882510416666666,
      "lo": 0.003882510416666666,
      "hi": 0.003882510416666666,
      "n": 72,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "country=MY",
      "target": "diy:gemini-3.8-flash",
      "value": 0.003882510416666666,
      "lo": 0.003882510416666666,
      "hi": 0.003882510416666666,
      "n": 72,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "country=MY",
      "target": "diy:gemini-3.8-flash",
      "value": 0.8048780487804879,
      "lo": 0.7613886230457927,
      "hi": 0.8420850376132984,
      "n": 369,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "country=MY",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.010303196519090837,
      "n": 369,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "country=MY",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.010303196519090837,
      "n": 369,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "currency=JPY",
      "target": "diy:gemini-3.8-flash",
      "value": 0.6666666666666666,
      "lo": 0.39062208887279953,
      "hi": 0.8618799089087867,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "currency=JPY",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "currency=JPY",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "currency=MYR",
      "target": "diy:gemini-3.8-flash",
      "value": 5981.5,
      "lo": 5981.5,
      "hi": 5981.5,
      "n": 72,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "currency=MYR",
      "target": "diy:gemini-3.8-flash",
      "value": 9944.100000000002,
      "lo": 9944.100000000002,
      "hi": 9944.100000000002,
      "n": 72,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "currency=MYR",
      "target": "diy:gemini-3.8-flash",
      "value": 20084.660000000054,
      "lo": 20084.660000000054,
      "hi": 20084.660000000054,
      "n": 72,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "currency=MYR",
      "target": "diy:gemini-3.8-flash",
      "value": 0.003882510416666666,
      "lo": 0.003882510416666666,
      "hi": 0.003882510416666666,
      "n": 72,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "currency=MYR",
      "target": "diy:gemini-3.8-flash",
      "value": 0.003882510416666666,
      "lo": 0.003882510416666666,
      "hi": 0.003882510416666666,
      "n": 72,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "currency=MYR",
      "target": "diy:gemini-3.8-flash",
      "value": 0.8048780487804879,
      "lo": 0.7613886230457927,
      "hi": 0.8420850376132984,
      "n": 369,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "currency=MYR",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.010303196519090837,
      "n": 369,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "currency=MYR",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.010303196519090837,
      "n": 369,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "dateEra=old",
      "target": "diy:gemini-3.8-flash",
      "value": 6259,
      "lo": 6259,
      "hi": 6259,
      "n": 77,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "dateEra=old",
      "target": "diy:gemini-3.8-flash",
      "value": 14472.000000000004,
      "lo": 14472.000000000004,
      "hi": 14472.000000000004,
      "n": 77,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "dateEra=old",
      "target": "diy:gemini-3.8-flash",
      "value": 27459.239999999972,
      "lo": 27459.239999999972,
      "hi": 27459.239999999972,
      "n": 77,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "dateEra=old",
      "target": "diy:gemini-3.8-flash",
      "value": 0.004113857142857141,
      "lo": 0.004113857142857141,
      "hi": 0.004113857142857141,
      "n": 77,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "dateEra=old",
      "target": "diy:gemini-3.8-flash",
      "value": 0.004113857142857141,
      "lo": 0.004113857142857141,
      "hi": 0.004113857142857141,
      "n": 77,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "dateEra=old",
      "target": "diy:gemini-3.8-flash",
      "value": 0.7994791666666666,
      "lo": 0.7565548645114459,
      "hi": 0.8364709576875495,
      "n": 384,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "dateEra=old",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 8.673617379884035e-19,
      "hi": 0.009904714241677031,
      "n": 384,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "dateEra=old",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 8.673617379884035e-19,
      "hi": 0.009904714241677031,
      "n": 384,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "dateEra=recent",
      "target": "diy:gemini-3.8-flash",
      "value": 8900,
      "lo": 8900,
      "hi": 8900,
      "n": 21,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "dateEra=recent",
      "target": "diy:gemini-3.8-flash",
      "value": 18099,
      "lo": 18099,
      "hi": 18099,
      "n": 21,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "dateEra=recent",
      "target": "diy:gemini-3.8-flash",
      "value": 20627.000000000004,
      "lo": 20627.000000000004,
      "hi": 20627.000000000004,
      "n": 21,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "dateEra=recent",
      "target": "diy:gemini-3.8-flash",
      "value": 0.005632035714285714,
      "lo": 0.005632035714285714,
      "hi": 0.005632035714285714,
      "n": 21,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "dateEra=recent",
      "target": "diy:gemini-3.8-flash",
      "value": 0.005632035714285714,
      "lo": 0.005632035714285714,
      "hi": 0.005632035714285714,
      "n": 21,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "dateEra=recent",
      "target": "diy:gemini-3.8-flash",
      "value": 0.6956521739130435,
      "lo": 0.5791642983213836,
      "hi": 0.7915037310829005,
      "n": 69,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "dateEra=recent",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 3.469446951953614e-18,
      "hi": 0.05273725818905173,
      "n": 69,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "dateEra=recent",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 3.469446951953614e-18,
      "hi": 0.05273725818905173,
      "n": 69,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "docType=receipt",
      "target": "diy:gemini-3.8-flash",
      "value": 6522.5,
      "lo": 6522.5,
      "hi": 6522.5,
      "n": 98,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "docType=receipt",
      "target": "diy:gemini-3.8-flash",
      "value": 16140.199999999986,
      "lo": 16140.199999999986,
      "hi": 16140.199999999986,
      "n": 98,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "docType=receipt",
      "target": "diy:gemini-3.8-flash",
      "value": 26382.780000000006,
      "lo": 26382.780000000006,
      "hi": 26382.780000000006,
      "n": 98,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "docType=receipt",
      "target": "diy:gemini-3.8-flash",
      "value": 0.004439181122448979,
      "lo": 0.004439181122448979,
      "hi": 0.004439181122448979,
      "n": 98,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "docType=receipt",
      "target": "diy:gemini-3.8-flash",
      "value": 0.004439181122448979,
      "lo": 0.004439181122448979,
      "hi": 0.004439181122448979,
      "n": 98,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "docType=receipt",
      "target": "diy:gemini-3.8-flash",
      "value": 0.7836644591611479,
      "lo": 0.7434472034386128,
      "hi": 0.8191111962815625,
      "n": 453,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "docType=receipt",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.008408735123582253,
      "n": 453,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "docType=receipt",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.008408735123582253,
      "n": 453,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "editedField=date",
      "target": "diy:gemini-3.8-flash",
      "value": 1,
      "lo": 0.7575059933447592,
      "hi": 1,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "editedField=date",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "editedField=date",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "editedField=items+total",
      "target": "diy:gemini-3.8-flash",
      "value": 1,
      "lo": 0.7575059933447592,
      "hi": 1,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "editedField=items+total",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "editedField=items+total",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "editedField=total",
      "target": "diy:gemini-3.8-flash",
      "value": 1,
      "lo": 0.8754449702581328,
      "hi": 0.9999999999999998,
      "n": 27,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "editedField=total",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.12455502974186705,
      "n": 27,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "editedField=total",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.12455502974186705,
      "n": 27,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "lang=en",
      "target": "diy:gemini-3.8-flash",
      "value": 6271.5,
      "lo": 6271.5,
      "hi": 6271.5,
      "n": 78,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "lang=en",
      "target": "diy:gemini-3.8-flash",
      "value": 11703.84999999996,
      "lo": 11703.84999999996,
      "hi": 11703.84999999996,
      "n": 78,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "lang=en",
      "target": "diy:gemini-3.8-flash",
      "value": 27407.98000000002,
      "lo": 27407.98000000002,
      "hi": 27407.98000000002,
      "n": 78,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "lang=en",
      "target": "diy:gemini-3.8-flash",
      "value": 0.004107038461538461,
      "lo": 0.004107038461538461,
      "hi": 0.004107038461538461,
      "n": 78,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "lang=en",
      "target": "diy:gemini-3.8-flash",
      "value": 0.004107038461538461,
      "lo": 0.004107038461538461,
      "hi": 0.004107038461538461,
      "n": 78,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "lang=en",
      "target": "diy:gemini-3.8-flash",
      "value": 0.7984496124031008,
      "lo": 0.755637501825676,
      "hi": 0.8353949866339876,
      "n": 387,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "lang=en",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.009828688164979105,
      "n": 387,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "lang=en",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.009828688164979105,
      "n": 387,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "lang=ja",
      "target": "diy:gemini-3.8-flash",
      "value": 0.6666666666666666,
      "lo": 0.39062208887279953,
      "hi": 0.8618799089087867,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "lang=ja",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "lang=ja",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "lang=zh",
      "target": "diy:gemini-3.8-flash",
      "value": 0.7222222222222222,
      "lo": 0.4912734343970254,
      "hi": 0.8750024662044601,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "lang=zh",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.17587922364665765,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "lang=zh",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.17587922364665765,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "prompt=explicit_amounts",
      "target": "diy:gemini-3.8-flash",
      "value": 0.7083333333333334,
      "lo": 0.5083230633550337,
      "hi": 0.8508535180303913,
      "n": 24,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "prompt=explicit_amounts",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 1.3877787807814457e-17,
      "hi": 0.13797620467498017,
      "n": 24,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "prompt=explicit_amounts",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 1.3877787807814457e-17,
      "hi": 0.13797620467498017,
      "n": 24,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "prompt=free",
      "target": "diy:gemini-3.8-flash",
      "value": 0.6666666666666666,
      "lo": 0.41713547738519774,
      "hi": 0.8482367556028526,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "prompt=free",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "prompt=free",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "prompt=target_total",
      "target": "diy:gemini-3.8-flash",
      "value": 0.7619047619047619,
      "lo": 0.5490882631704924,
      "hi": 0.8937198697736354,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "prompt=target_total",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.154639018924847,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "prompt=target_total",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.154639018924847,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "provenance=intact",
      "target": "diy:gemini-3.8-flash",
      "value": 9741.5,
      "lo": 9741.5,
      "hi": 9741.5,
      "n": 10,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "provenance=intact",
      "target": "diy:gemini-3.8-flash",
      "value": 29048.299999999996,
      "lo": 29048.299999999996,
      "hi": 29048.299999999996,
      "n": 10,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "provenance=intact",
      "target": "diy:gemini-3.8-flash",
      "value": 30893.66,
      "lo": 30893.66,
      "hi": 30893.66,
      "n": 10,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "provenance=intact",
      "target": "diy:gemini-3.8-flash",
      "value": 0.00712425,
      "lo": 0.00712425,
      "hi": 0.00712425,
      "n": 10,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "provenance=intact",
      "target": "diy:gemini-3.8-flash",
      "value": 0.00712425,
      "lo": 0.00712425,
      "hi": 0.00712425,
      "n": 10,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "provenance=intact",
      "target": "diy:gemini-3.8-flash",
      "value": 0.8333333333333334,
      "lo": 0.7196838683638547,
      "hi": 0.9068682302080855,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "provenance=intact",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.06017185214208986,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "provenance=intact",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.06017185214208986,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "provenance=none",
      "target": "diy:gemini-3.8-flash",
      "value": 6001,
      "lo": 6001,
      "hi": 6001,
      "n": 81,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "provenance=none",
      "target": "diy:gemini-3.8-flash",
      "value": 10548,
      "lo": 10548,
      "hi": 10548,
      "n": 81,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "provenance=none",
      "target": "diy:gemini-3.8-flash",
      "value": 16783.000000000015,
      "lo": 16783.000000000015,
      "hi": 16783.000000000015,
      "n": 81,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "provenance=none",
      "target": "diy:gemini-3.8-flash",
      "value": 0.00385837962962963,
      "lo": 0.00385837962962963,
      "hi": 0.00385837962962963,
      "n": 81,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "provenance=none",
      "target": "diy:gemini-3.8-flash",
      "value": 0.00385837962962963,
      "lo": 0.00385837962962963,
      "hi": 0.00385837962962963,
      "n": 81,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "provenance=none",
      "target": "diy:gemini-3.8-flash",
      "value": 0.7567567567567568,
      "lo": 0.7079173792646796,
      "hi": 0.7997398449935313,
      "n": 333,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "provenance=none",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.011404352760326315,
      "n": 333,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "provenance=none",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.011404352760326315,
      "n": 333,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "provenance=stripped",
      "target": "diy:gemini-3.8-flash",
      "value": 0.8833333333333333,
      "lo": 0.7782121197582914,
      "hi": 0.9423227935994396,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "provenance=stripped",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.06017185214208986,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "provenance=stripped",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.06017185214208986,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "source=commons",
      "target": "diy:gemini-3.8-flash",
      "value": 6632,
      "lo": 6632,
      "hi": 6632,
      "n": 13,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "source=commons",
      "target": "diy:gemini-3.8-flash",
      "value": 15510.999999999985,
      "lo": 15510.999999999985,
      "hi": 15510.999999999985,
      "n": 13,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "source=commons",
      "target": "diy:gemini-3.8-flash",
      "value": 20109.39999999999,
      "lo": 20109.39999999999,
      "hi": 20109.39999999999,
      "n": 13,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "source=commons",
      "target": "diy:gemini-3.8-flash",
      "value": 0.004866634615384616,
      "lo": 0.004866634615384616,
      "hi": 0.004866634615384616,
      "n": 13,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "source=commons",
      "target": "diy:gemini-3.8-flash",
      "value": 0.004866634615384616,
      "lo": 0.004866634615384616,
      "hi": 0.004866634615384616,
      "n": 13,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "source=commons",
      "target": "diy:gemini-3.8-flash",
      "value": 0.6666666666666666,
      "lo": 0.5097739439934599,
      "hi": 0.7936704381040163,
      "n": 39,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "source=commons",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.08966685370757144,
      "n": 39,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "source=commons",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.08966685370757144,
      "n": 39,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "source=edited:gemini-nano-banana-2.1",
      "target": "diy:gemini-3.8-flash",
      "value": 1,
      "lo": 0.7575059933447592,
      "hi": 1,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "source=edited:gemini-nano-banana-2.1",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "source=edited:gemini-nano-banana-2.1",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "source=edited:gpt-image-2.5-sunburst",
      "target": "diy:gemini-3.8-flash",
      "value": 1,
      "lo": 0.7575059933447592,
      "hi": 1,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "source=edited:gpt-image-2.5-sunburst",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "source=edited:gpt-image-2.5-sunburst",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "source=edited:sharp-patch",
      "target": "diy:gemini-3.8-flash",
      "value": 1,
      "lo": 0.8864866068260312,
      "hi": 0.9999999999999999,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "source=edited:sharp-patch",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "source=edited:sharp-patch",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "source=generated:flux-3-image",
      "target": "diy:gemini-3.8-flash",
      "value": 0.8333333333333334,
      "lo": 0.6077796189608428,
      "hi": 0.9416342319413856,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "source=generated:flux-3-image",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.17587922364665765,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "source=generated:flux-3-image",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.17587922364665765,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "source=generated:gemini-nano-banana-2.1",
      "target": "diy:gemini-3.8-flash",
      "value": 0.6666666666666666,
      "lo": 0.4537345197425524,
      "hi": 0.8280524739491653,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "source=generated:gemini-nano-banana-2.1",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.154639018924847,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "source=generated:gemini-nano-banana-2.1",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.154639018924847,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "source=generated:gpt-image-2.5-flare",
      "target": "diy:gemini-3.8-flash",
      "value": 0.6666666666666666,
      "lo": 0.4537345197425524,
      "hi": 0.8280524739491653,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "source=generated:gpt-image-2.5-flare",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.154639018924847,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "source=generated:gpt-image-2.5-flare",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.154639018924847,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "source=sroie",
      "target": "diy:gemini-3.8-flash",
      "value": 5624,
      "lo": 5624,
      "hi": 5624,
      "n": 68,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "source=sroie",
      "target": "diy:gemini-3.8-flash",
      "value": 9042.85,
      "lo": 9042.85,
      "hi": 9042.85,
      "n": 68,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "source=sroie",
      "target": "diy:gemini-3.8-flash",
      "value": 12236.279999999992,
      "lo": 12236.279999999992,
      "hi": 12236.279999999992,
      "n": 68,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "source=sroie",
      "target": "diy:gemini-3.8-flash",
      "value": 0.0036656249999999996,
      "lo": 0.0036656249999999996,
      "hi": 0.0036656249999999996,
      "n": 68,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "source=sroie",
      "target": "diy:gemini-3.8-flash",
      "value": 0.0036656249999999996,
      "lo": 0.0036656249999999996,
      "hi": 0.0036656249999999996,
      "n": 68,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "source=sroie",
      "target": "diy:gemini-3.8-flash",
      "value": 0.7687074829931972,
      "lo": 0.7172296222341344,
      "hi": 0.8132539465262468,
      "n": 294,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "source=sroie",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.012897663192708011,
      "n": 294,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "source=sroie",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.012897663192708011,
      "n": 294,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "transform=crop",
      "target": "diy:gemini-3.8-flash",
      "value": 1,
      "lo": 0.7961166989641515,
      "hi": 1,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "transform=crop",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "transform=crop",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "transform=jpeg",
      "target": "diy:gemini-3.8-flash",
      "value": 1,
      "lo": 0.7961166989641515,
      "hi": 1,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "transform=jpeg",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "transform=jpeg",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "transform=rephoto",
      "target": "diy:gemini-3.8-flash",
      "value": 1,
      "lo": 0.7961166989641515,
      "hi": 1,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "transform=rephoto",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "transform=rephoto",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "transform=resize",
      "target": "diy:gemini-3.8-flash",
      "value": 1,
      "lo": 0.7961166989641515,
      "hi": 1,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "transform=resize",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "transform=resize",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "transform=rotate",
      "target": "diy:gemini-3.8-flash",
      "value": 1,
      "lo": 0.7961166989641515,
      "hi": 1,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "transform=rotate",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "transform=rotate",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "transform=screenshot",
      "target": "diy:gemini-3.8-flash",
      "value": 1,
      "lo": 0.7961166989641515,
      "hi": 1,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "transform=screenshot",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "transform=screenshot",
      "target": "diy:gemini-3.8-flash",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "all",
      "target": "decision:gpt-6-luna",
      "value": 0.6479938271604938,
      "lo": 0.5361111111111111,
      "hi": 0.7547878086419754,
      "n": 121,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "all",
      "target": "decision:gpt-6-luna",
      "value": 0.25,
      "lo": 0.125,
      "hi": 0.425,
      "n": 40,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "all",
      "target": "decision:gpt-6-luna",
      "value": 0.3,
      "lo": 0.175,
      "hi": 0.475,
      "n": 40,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "all",
      "target": "decision:gpt-6-luna",
      "value": 0.35,
      "lo": 0.225,
      "hi": 0.55,
      "n": 40,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "all",
      "target": "decision:gpt-6-luna",
      "value": 0.225,
      "lo": 0.12316091323540743,
      "hi": 0.37503096742310027,
      "n": 40,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_5",
      "stratum": "all",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.045278085432415204,
      "n": 81,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "all",
      "target": "decision:gpt-6-luna",
      "value": 0.225,
      "lo": 0.12316091323540743,
      "hi": 0.37503096742310027,
      "n": 40,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_7",
      "stratum": "all",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.045278085432415204,
      "n": 81,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "base=sroie",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.16112515805281938,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "base=sroie",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.16112515805281938,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "capture=photo",
      "target": "decision:gpt-6-luna",
      "value": 0.8333333333333334,
      "lo": 0.6388888888888888,
      "hi": 0.9722222222222222,
      "n": 21,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "capture=photo",
      "target": "decision:gpt-6-luna",
      "value": 0.6666666666666666,
      "lo": 0.4166666666666667,
      "hi": 0.9166666666666666,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "capture=photo",
      "target": "decision:gpt-6-luna",
      "value": 0.6666666666666666,
      "lo": 0.4166666666666667,
      "hi": 0.9166666666666666,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "capture=photo",
      "target": "decision:gpt-6-luna",
      "value": 0.6666666666666666,
      "lo": 0.4166666666666667,
      "hi": 0.9166666666666666,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "capture=photo",
      "target": "decision:gpt-6-luna",
      "value": 0.5833333333333334,
      "lo": 0.31951131254954973,
      "hi": 0.8067396863412435,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "capture=photo",
      "target": "decision:gpt-6-luna",
      "value": 0.5833333333333334,
      "lo": 0.31951131254954973,
      "hi": 0.8067396863412435,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "capture=scan",
      "target": "decision:gpt-6-luna",
      "value": 0.5870221327967807,
      "lo": 0.44893737424547286,
      "hi": 0.7190769617706237,
      "n": 99,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "capture=scan",
      "target": "decision:gpt-6-luna",
      "value": 0.10714285714285714,
      "lo": 0,
      "hi": 0.2857142857142857,
      "n": 28,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "capture=scan",
      "target": "decision:gpt-6-luna",
      "value": 0.17857142857142858,
      "lo": 0.03571428571428571,
      "hi": 0.35714285714285715,
      "n": 28,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "capture=scan",
      "target": "decision:gpt-6-luna",
      "value": 0.25,
      "lo": 0.07142857142857142,
      "hi": 0.4294642857142808,
      "n": 28,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "capture=scan",
      "target": "decision:gpt-6-luna",
      "value": 0.07142857142857142,
      "lo": 0.019812064115305755,
      "hi": 0.22645362568399052,
      "n": 28,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_5",
      "stratum": "capture=scan",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.05132795219689034,
      "n": 71,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "capture=scan",
      "target": "decision:gpt-6-luna",
      "value": 0.07142857142857142,
      "lo": 0.019812064115305755,
      "hi": 0.22645362568399052,
      "n": 28,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_7",
      "stratum": "capture=scan",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.05132795219689034,
      "n": 71,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "country=MY",
      "target": "decision:gpt-6-luna",
      "value": 0.5258823529411765,
      "lo": 0.37969117647058825,
      "hi": 0.6655882352941176,
      "n": 93,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "country=MY",
      "target": "decision:gpt-6-luna",
      "value": 0.04,
      "lo": 0,
      "hi": 0.24,
      "n": 25,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "country=MY",
      "target": "decision:gpt-6-luna",
      "value": 0.12,
      "lo": 0,
      "hi": 0.36,
      "n": 25,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "country=MY",
      "target": "decision:gpt-6-luna",
      "value": 0.2,
      "lo": 0.04,
      "hi": 0.4,
      "n": 25,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "country=MY",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 1.3877787807814457e-17,
      "hi": 0.13319225093904846,
      "n": 25,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_5",
      "stratum": "country=MY",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.053471336520070537,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "country=MY",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 1.3877787807814457e-17,
      "hi": 0.13319225093904846,
      "n": 25,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_7",
      "stratum": "country=MY",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.053471336520070537,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "currency=MYR",
      "target": "decision:gpt-6-luna",
      "value": 0.5258823529411765,
      "lo": 0.3796691176470589,
      "hi": 0.6647132352941176,
      "n": 93,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "currency=MYR",
      "target": "decision:gpt-6-luna",
      "value": 0.04,
      "lo": 0,
      "hi": 0.24,
      "n": 25,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "currency=MYR",
      "target": "decision:gpt-6-luna",
      "value": 0.12,
      "lo": 0,
      "hi": 0.32,
      "n": 25,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "currency=MYR",
      "target": "decision:gpt-6-luna",
      "value": 0.2,
      "lo": 0.04,
      "hi": 0.4,
      "n": 25,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "currency=MYR",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 1.3877787807814457e-17,
      "hi": 0.13319225093904846,
      "n": 25,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_5",
      "stratum": "currency=MYR",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.053471336520070537,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "currency=MYR",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 1.3877787807814457e-17,
      "hi": 0.13319225093904846,
      "n": 25,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_7",
      "stratum": "currency=MYR",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.053471336520070537,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "dateEra=old",
      "target": "decision:gpt-6-luna",
      "value": 0.5482843137254902,
      "lo": 0.4124877450980392,
      "hi": 0.6806678921568626,
      "n": 98,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "dateEra=old",
      "target": "decision:gpt-6-luna",
      "value": 0.06666666666666667,
      "lo": 0,
      "hi": 0.23333333333333334,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "dateEra=old",
      "target": "decision:gpt-6-luna",
      "value": 0.13333333333333333,
      "lo": 0.03333333333333333,
      "hi": 0.3333333333333333,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "dateEra=old",
      "target": "decision:gpt-6-luna",
      "value": 0.2,
      "lo": 0.06666666666666667,
      "hi": 0.4,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "dateEra=old",
      "target": "decision:gpt-6-luna",
      "value": 0.03333333333333333,
      "lo": 0.005908590381612455,
      "hi": 0.16670390991409173,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_5",
      "stratum": "dateEra=old",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.053471336520070537,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "dateEra=old",
      "target": "decision:gpt-6-luna",
      "value": 0.03333333333333333,
      "lo": 0.005908590381612455,
      "hi": 0.16670390991409173,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_7",
      "stratum": "dateEra=old",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.053471336520070537,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "dateEra=recent",
      "target": "decision:gpt-6-luna",
      "value": 0.9230769230769231,
      "lo": 0.7692307692307693,
      "hi": 1,
      "n": 23,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "dateEra=recent",
      "target": "decision:gpt-6-luna",
      "value": 0.8,
      "lo": 0.5,
      "hi": 1,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "dateEra=recent",
      "target": "decision:gpt-6-luna",
      "value": 0.8,
      "lo": 0.5,
      "hi": 1,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "dateEra=recent",
      "target": "decision:gpt-6-luna",
      "value": 0.8,
      "lo": 0.5,
      "hi": 1,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "dateEra=recent",
      "target": "decision:gpt-6-luna",
      "value": 0.8,
      "lo": 0.49016247153664183,
      "hi": 0.9433178485456247,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_5",
      "stratum": "dateEra=recent",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.22809537235419838,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "dateEra=recent",
      "target": "decision:gpt-6-luna",
      "value": 0.8,
      "lo": 0.49016247153664183,
      "hi": 0.9433178485456247,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_7",
      "stratum": "dateEra=recent",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.22809537235419838,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "docType=receipt",
      "target": "decision:gpt-6-luna",
      "value": 0.6479938271604938,
      "lo": 0.5340779320987655,
      "hi": 0.757739197530864,
      "n": 121,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "docType=receipt",
      "target": "decision:gpt-6-luna",
      "value": 0.25,
      "lo": 0.15,
      "hi": 0.425,
      "n": 40,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "docType=receipt",
      "target": "decision:gpt-6-luna",
      "value": 0.3,
      "lo": 0.175,
      "hi": 0.475,
      "n": 40,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "docType=receipt",
      "target": "decision:gpt-6-luna",
      "value": 0.35,
      "lo": 0.2,
      "hi": 0.55,
      "n": 40,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "docType=receipt",
      "target": "decision:gpt-6-luna",
      "value": 0.225,
      "lo": 0.12316091323540743,
      "hi": 0.37503096742310027,
      "n": 40,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_5",
      "stratum": "docType=receipt",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.045278085432415204,
      "n": 81,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "docType=receipt",
      "target": "decision:gpt-6-luna",
      "value": 0.225,
      "lo": 0.12316091323540743,
      "hi": 0.37503096742310027,
      "n": 40,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_7",
      "stratum": "docType=receipt",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.045278085432415204,
      "n": 81,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "lang=en",
      "target": "decision:gpt-6-luna",
      "value": 0.5785024154589372,
      "lo": 0.4438405797101451,
      "hi": 0.7091847826086956,
      "n": 99,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "lang=en",
      "target": "decision:gpt-6-luna",
      "value": 0.13333333333333333,
      "lo": 0.03333333333333333,
      "hi": 0.3333333333333333,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "lang=en",
      "target": "decision:gpt-6-luna",
      "value": 0.2,
      "lo": 0.06666666666666667,
      "hi": 0.4,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "lang=en",
      "target": "decision:gpt-6-luna",
      "value": 0.26666666666666666,
      "lo": 0.1,
      "hi": 0.4666666666666667,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "lang=en",
      "target": "decision:gpt-6-luna",
      "value": 0.1,
      "lo": 0.03459988874733419,
      "hi": 0.25621082579184085,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_5",
      "stratum": "lang=en",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 3.469446951953614e-18,
      "hi": 0.05273725818905173,
      "n": 69,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "lang=en",
      "target": "decision:gpt-6-luna",
      "value": 0.1,
      "lo": 0.03459988874733419,
      "hi": 0.25621082579184085,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_7",
      "stratum": "lang=en",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 3.469446951953614e-18,
      "hi": 0.05273725818905173,
      "n": 69,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_5",
      "stratum": "source=commons",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.22809537235419838,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_7",
      "stratum": "source=commons",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.22809537235419838,
      "n": 13,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "source=edited:sharp-patch",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.2775327998628892,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "source=edited:sharp-patch",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.2775327998628892,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_5",
      "stratum": "source=sroie",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.053471336520070537,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_7",
      "stratum": "source=sroie",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.053471336520070537,
      "n": 68,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "provenance=intact",
      "target": "decision:gpt-6-luna",
      "value": 0.25,
      "lo": 0.11186170140766569,
      "hi": 0.468700877618744,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "provenance=intact",
      "target": "decision:gpt-6-luna",
      "value": 0.25,
      "lo": 0.11186170140766569,
      "hi": 0.468700877618744,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "provenance=stripped",
      "target": "decision:gpt-6-luna",
      "value": 0.2,
      "lo": 0.08065766257979806,
      "hi": 0.41601743225189364,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "provenance=stripped",
      "target": "decision:gpt-6-luna",
      "value": 0.2,
      "lo": 0.08065766257979806,
      "hi": 0.41601743225189364,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_5",
      "stratum": "provenance=none",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.045278085432415204,
      "n": 81,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "false_flag_rate_at_risk_0_7",
      "stratum": "provenance=none",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.045278085432415204,
      "n": 81,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "class=ai_full",
      "target": "decision:gpt-6-luna",
      "value": 0.7280864197530864,
      "lo": 0.5681867283950618,
      "hi": 0.8651311728395061,
      "n": 101,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "class=ai_full",
      "target": "decision:gpt-6-luna",
      "value": 0.45,
      "lo": 0.25,
      "hi": 0.6512499999999932,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "class=ai_full",
      "target": "decision:gpt-6-luna",
      "value": 0.45,
      "lo": 0.25,
      "hi": 0.7,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "class=ai_full",
      "target": "decision:gpt-6-luna",
      "value": 0.5,
      "lo": 0.3,
      "hi": 0.7,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "class=ai_full",
      "target": "decision:gpt-6-luna",
      "value": 0.45,
      "lo": 0.2581978582556243,
      "hi": 0.6579146575496577,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "class=ai_full",
      "target": "decision:gpt-6-luna",
      "value": 0.45,
      "lo": 0.2581978582556243,
      "hi": 0.6579146575496577,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "class=ai_inpaint",
      "target": "decision:gpt-6-luna",
      "value": 0.6407407407407407,
      "lo": 0.44012345679012344,
      "hi": 0.8327314814814814,
      "n": 91,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "class=ai_inpaint",
      "target": "decision:gpt-6-luna",
      "value": 0.1,
      "lo": 0,
      "hi": 0.4,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "class=ai_inpaint",
      "target": "decision:gpt-6-luna",
      "value": 0.2,
      "lo": 0,
      "hi": 0.6,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "class=ai_inpaint",
      "target": "decision:gpt-6-luna",
      "value": 0.3,
      "lo": 0,
      "hi": 0.6,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "class=ai_inpaint",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.2775327998628892,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "class=ai_inpaint",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.2775327998628892,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "class=manual_edit",
      "target": "decision:gpt-6-luna",
      "value": 0.49506172839506174,
      "lo": 0.3012191358024692,
      "hi": 0.6858179012345678,
      "n": 91,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "class=manual_edit",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.3,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "class=manual_edit",
      "target": "decision:gpt-6-luna",
      "value": 0.1,
      "lo": 0,
      "hi": 0.3,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "class=manual_edit",
      "target": "decision:gpt-6-luna",
      "value": 0.1,
      "lo": 0,
      "hi": 0.5,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "class=manual_edit",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.2775327998628892,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "class=manual_edit",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.2775327998628892,
      "n": 10,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "arithmetic=consistent",
      "target": "decision:gpt-6-luna",
      "value": 0.6554433221099888,
      "lo": 0.5193602693602694,
      "hi": 0.7895693041526375,
      "n": 103,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "arithmetic=consistent",
      "target": "decision:gpt-6-luna",
      "value": 0.2727272727272727,
      "lo": 0.09090909090909091,
      "hi": 0.45454545454545453,
      "n": 22,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "arithmetic=consistent",
      "target": "decision:gpt-6-luna",
      "value": 0.2727272727272727,
      "lo": 0.09090909090909091,
      "hi": 0.45454545454545453,
      "n": 22,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "arithmetic=consistent",
      "target": "decision:gpt-6-luna",
      "value": 0.2727272727272727,
      "lo": 0.13636363636363635,
      "hi": 0.5,
      "n": 22,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "arithmetic=consistent",
      "target": "decision:gpt-6-luna",
      "value": 0.2727272727272727,
      "lo": 0.13150760287989469,
      "hi": 0.4815173401240238,
      "n": 22,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "arithmetic=consistent",
      "target": "decision:gpt-6-luna",
      "value": 0.2727272727272727,
      "lo": 0.13150760287989469,
      "hi": 0.4815173401240238,
      "n": 22,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "auc",
      "stratum": "arithmetic=inconsistent",
      "target": "decision:gpt-6-luna",
      "value": 0.6388888888888888,
      "lo": 0.4639574759945131,
      "hi": 0.8079646776406034,
      "n": 99,
      "unit": "auc",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_1",
      "stratum": "arithmetic=inconsistent",
      "target": "decision:gpt-6-luna",
      "value": 0.2222222222222222,
      "lo": 0.05555555555555555,
      "hi": 0.5,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_5",
      "stratum": "arithmetic=inconsistent",
      "target": "decision:gpt-6-luna",
      "value": 0.3333333333333333,
      "lo": 0.16666666666666666,
      "hi": 0.6111111111111112,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_fpr_10",
      "stratum": "arithmetic=inconsistent",
      "target": "decision:gpt-6-luna",
      "value": 0.4444444444444444,
      "lo": 0.2222222222222222,
      "hi": 0.6666666666666666,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_5",
      "stratum": "arithmetic=inconsistent",
      "target": "decision:gpt-6-luna",
      "value": 0.16666666666666666,
      "lo": 0.05836576805861446,
      "hi": 0.39222038103915724,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "authenticity",
      "metric": "detection_at_risk_0_7",
      "stratum": "arithmetic=inconsistent",
      "target": "decision:gpt-6-luna",
      "value": 0.16666666666666666,
      "lo": 0.05836576805861446,
      "hi": 0.39222038103915724,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "decisions",
      "metric": "decision_consistency",
      "stratum": "all",
      "target": "decision:gpt-6-luna",
      "value": 1,
      "lo": 0.9692293020525218,
      "hi": 1,
      "n": 121,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "decisions",
      "metric": "total_consistency",
      "stratum": "all",
      "target": "decision:gpt-6-luna",
      "value": 1,
      "lo": 0.9692293020525218,
      "hi": 1,
      "n": 121,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "decisions",
      "metric": "calibration_error",
      "stratum": "all",
      "target": "decision:gpt-6-luna",
      "value": 0.18347107438016522,
      "lo": 0.14495661157024786,
      "hi": 0.2177727272727272,
      "n": 121,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_5",
      "stratum": "all",
      "target": "decision:gpt-6-luna",
      "value": 775.0000000000001,
      "lo": 625,
      "hi": 900.0000000000001,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_7",
      "stratum": "all",
      "target": "decision:gpt-6-luna",
      "value": 775.0000000000001,
      "lo": 650,
      "hi": 900.0000000000001,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_5",
      "stratum": "assumption=fakes_0_1pct",
      "target": "decision:gpt-6-luna",
      "value": 77.50000000000001,
      "lo": 65,
      "hi": 90.00000000000001,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_7",
      "stratum": "assumption=fakes_0_1pct",
      "target": "decision:gpt-6-luna",
      "value": 77.50000000000001,
      "lo": 65,
      "hi": 90.00000000000001,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_5",
      "stratum": "assumption=fakes_5pct",
      "target": "decision:gpt-6-luna",
      "value": 3875.000000000001,
      "lo": 3250,
      "hi": 4500.000000000001,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_7",
      "stratum": "assumption=fakes_5pct",
      "target": "decision:gpt-6-luna",
      "value": 3875.000000000001,
      "lo": 3250,
      "hi": 4500.000000000001,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_5",
      "stratum": "assumption=fake_32usd",
      "target": "decision:gpt-6-luna",
      "value": 248.00000000000003,
      "lo": 207.80000000000038,
      "hi": 288.00000000000006,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_7",
      "stratum": "assumption=fake_32usd",
      "target": "decision:gpt-6-luna",
      "value": 248.00000000000003,
      "lo": 208.00000000000003,
      "hi": 288.00000000000006,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_5",
      "stratum": "assumption=fake_182usd",
      "target": "decision:gpt-6-luna",
      "value": 1410.5,
      "lo": 1183,
      "hi": 1638.0000000000002,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_7",
      "stratum": "assumption=fake_182usd",
      "target": "decision:gpt-6-luna",
      "value": 1410.5,
      "lo": 1183,
      "hi": 1638.0000000000002,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_5",
      "stratum": "assumption=review_8usd",
      "target": "decision:gpt-6-luna",
      "value": 775.0000000000001,
      "lo": 650,
      "hi": 900.0000000000001,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_7",
      "stratum": "assumption=review_8usd",
      "target": "decision:gpt-6-luna",
      "value": 775.0000000000001,
      "lo": 650,
      "hi": 900.0000000000001,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_5",
      "stratum": "assumption=review_73usd",
      "target": "decision:gpt-6-luna",
      "value": 775.0000000000001,
      "lo": 650,
      "hi": 900.0000000000001,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "expected_cost_per_1000_at_risk_0_7",
      "stratum": "assumption=review_73usd",
      "target": "decision:gpt-6-luna",
      "value": 775.0000000000001,
      "lo": 650,
      "hi": 900.0000000000001,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "approve_all_cost_per_1000",
      "stratum": "all",
      "target": "decision:gpt-6-luna",
      "value": 1000,
      "lo": 1000,
      "hi": 1000,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "approve_all_cost_per_1000",
      "stratum": "assumption=fakes_0_1pct",
      "target": "decision:gpt-6-luna",
      "value": 100,
      "lo": 100,
      "hi": 100,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "approve_all_cost_per_1000",
      "stratum": "assumption=fakes_5pct",
      "target": "decision:gpt-6-luna",
      "value": 5000,
      "lo": 5000,
      "hi": 5000,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "approve_all_cost_per_1000",
      "stratum": "assumption=fake_32usd",
      "target": "decision:gpt-6-luna",
      "value": 320,
      "lo": 320,
      "hi": 320,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "approve_all_cost_per_1000",
      "stratum": "assumption=fake_182usd",
      "target": "decision:gpt-6-luna",
      "value": 1820,
      "lo": 1820,
      "hi": 1820,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "approve_all_cost_per_1000",
      "stratum": "assumption=review_8usd",
      "target": "decision:gpt-6-luna",
      "value": 1000,
      "lo": 1000,
      "hi": 1000,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "decisions",
      "metric": "approve_all_cost_per_1000",
      "stratum": "assumption=review_73usd",
      "target": "decision:gpt-6-luna",
      "value": 1000,
      "lo": 1000,
      "hi": 1000,
      "n": 121,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "evidence",
      "metric": "edited_field_named",
      "stratum": "all",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.16112515805281938,
      "n": 20,
      "unit": "rate",
      "higherIsBetter": true
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "all",
      "target": "decision:gpt-6-luna",
      "value": 349,
      "lo": 349,
      "hi": 349,
      "n": 453,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "all",
      "target": "decision:gpt-6-luna",
      "value": 565.7999999999995,
      "lo": 565.7999999999995,
      "hi": 565.7999999999995,
      "n": 453,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "all",
      "target": "decision:gpt-6-luna",
      "value": 829.48,
      "lo": 829.48,
      "hi": 829.48,
      "n": 453,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "all",
      "target": "decision:gpt-6-luna",
      "value": 0.0001681125827814572,
      "lo": 0.0001681125827814572,
      "hi": 0.0001681125827814572,
      "n": 453,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "all",
      "target": "decision:gpt-6-luna",
      "value": 0.0001681125827814572,
      "lo": 0.0001681125827814572,
      "hi": 0.0001681125827814572,
      "n": 453,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "all",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.008408735123582253,
      "n": 453,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "all",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.008408735123582253,
      "n": 453,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "all",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.008408735123582253,
      "n": 453,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "arithmetic=consistent",
      "target": "decision:gpt-6-luna",
      "value": 351.5,
      "lo": 351.5,
      "hi": 351.5,
      "n": 390,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "arithmetic=consistent",
      "target": "decision:gpt-6-luna",
      "value": 569.099999999999,
      "lo": 569.099999999999,
      "hi": 569.099999999999,
      "n": 390,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "arithmetic=consistent",
      "target": "decision:gpt-6-luna",
      "value": 831.5400000000002,
      "lo": 831.5400000000002,
      "hi": 831.5400000000002,
      "n": 390,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "arithmetic=consistent",
      "target": "decision:gpt-6-luna",
      "value": 0.00016913846153846185,
      "lo": 0.00016913846153846185,
      "hi": 0.00016913846153846185,
      "n": 390,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "arithmetic=consistent",
      "target": "decision:gpt-6-luna",
      "value": 0.00016913846153846185,
      "lo": 0.00016913846153846185,
      "hi": 0.00016913846153846185,
      "n": 390,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "arithmetic=consistent",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.009753820311850517,
      "n": 390,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "arithmetic=consistent",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.009753820311850517,
      "n": 390,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "arithmetic=consistent",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.009753820311850517,
      "n": 390,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "arithmetic=inconsistent",
      "target": "decision:gpt-6-luna",
      "value": 342.5,
      "lo": 342.5,
      "hi": 342.5,
      "n": 54,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "arithmetic=inconsistent",
      "target": "decision:gpt-6-luna",
      "value": 558.3499999999997,
      "lo": 558.3499999999997,
      "hi": 558.3499999999997,
      "n": 54,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "arithmetic=inconsistent",
      "target": "decision:gpt-6-luna",
      "value": 653.87,
      "lo": 653.87,
      "hi": 653.87,
      "n": 54,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "arithmetic=inconsistent",
      "target": "decision:gpt-6-luna",
      "value": 0.00016527777777777783,
      "lo": 0.00016527777777777783,
      "hi": 0.00016527777777777783,
      "n": 54,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "arithmetic=inconsistent",
      "target": "decision:gpt-6-luna",
      "value": 0.00016527777777777783,
      "lo": 0.00016527777777777783,
      "hi": 0.00016527777777777783,
      "n": 54,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "arithmetic=inconsistent",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.0664135880908964,
      "n": 54,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "arithmetic=inconsistent",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.0664135880908964,
      "n": 54,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "arithmetic=inconsistent",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.0664135880908964,
      "n": 54,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "arithmeticError=totals",
      "target": "decision:gpt-6-luna",
      "value": 374,
      "lo": 374,
      "hi": 374,
      "n": 21,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "arithmeticError=totals",
      "target": "decision:gpt-6-luna",
      "value": 537,
      "lo": 537,
      "hi": 537,
      "n": 21,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "arithmeticError=totals",
      "target": "decision:gpt-6-luna",
      "value": 585.8000000000001,
      "lo": 585.8000000000001,
      "hi": 585.8000000000001,
      "n": 21,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "arithmeticError=totals",
      "target": "decision:gpt-6-luna",
      "value": 0.0001725714285714285,
      "lo": 0.0001725714285714285,
      "hi": 0.0001725714285714285,
      "n": 21,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "arithmeticError=totals",
      "target": "decision:gpt-6-luna",
      "value": 0.0001725714285714285,
      "lo": 0.0001725714285714285,
      "hi": 0.0001725714285714285,
      "n": 21,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "arithmeticError=totals",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.154639018924847,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "arithmeticError=totals",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.154639018924847,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "arithmeticError=totals",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.154639018924847,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "base=sroie",
      "target": "decision:gpt-6-luna",
      "value": 342.5,
      "lo": 342.5,
      "hi": 342.5,
      "n": 60,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "base=sroie",
      "target": "decision:gpt-6-luna",
      "value": 478.24999999999994,
      "lo": 478.24999999999994,
      "hi": 478.24999999999994,
      "n": 60,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "base=sroie",
      "target": "decision:gpt-6-luna",
      "value": 652.6099999999999,
      "lo": 652.6099999999999,
      "hi": 652.6099999999999,
      "n": 60,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "base=sroie",
      "target": "decision:gpt-6-luna",
      "value": 0.00016975,
      "lo": 0.00016975,
      "hi": 0.00016975,
      "n": 60,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "base=sroie",
      "target": "decision:gpt-6-luna",
      "value": 0.00016975,
      "lo": 0.00016975,
      "hi": 0.00016975,
      "n": 60,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "base=sroie",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.06017185214208986,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "base=sroie",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.06017185214208986,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "base=sroie",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.06017185214208986,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "capture=photo",
      "target": "decision:gpt-6-luna",
      "value": 380,
      "lo": 380,
      "hi": 380,
      "n": 75,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "capture=photo",
      "target": "decision:gpt-6-luna",
      "value": 638.9,
      "lo": 638.9,
      "hi": 638.9,
      "n": 75,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "capture=photo",
      "target": "decision:gpt-6-luna",
      "value": 799.6000000000004,
      "lo": 799.6000000000004,
      "hi": 799.6000000000004,
      "n": 75,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "capture=photo",
      "target": "decision:gpt-6-luna",
      "value": 0.0001894800000000001,
      "lo": 0.0001894800000000001,
      "hi": 0.0001894800000000001,
      "n": 75,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "capture=photo",
      "target": "decision:gpt-6-luna",
      "value": 0.0001894800000000001,
      "lo": 0.0001894800000000001,
      "hi": 0.0001894800000000001,
      "n": 75,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "capture=photo",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 3.469446951953614e-18,
      "hi": 0.04872384248280587,
      "n": 75,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "capture=photo",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 3.469446951953614e-18,
      "hi": 0.04872384248280587,
      "n": 75,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "capture=photo",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 3.469446951953614e-18,
      "hi": 0.04872384248280587,
      "n": 75,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "capture=scan",
      "target": "decision:gpt-6-luna",
      "value": 342,
      "lo": 342,
      "hi": 342,
      "n": 360,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "capture=scan",
      "target": "decision:gpt-6-luna",
      "value": 502.4000000000001,
      "lo": 502.4000000000001,
      "hi": 502.4000000000001,
      "n": 360,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "capture=scan",
      "target": "decision:gpt-6-luna",
      "value": 774.0700000000006,
      "lo": 774.0700000000006,
      "hi": 774.0700000000006,
      "n": 360,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "capture=scan",
      "target": "decision:gpt-6-luna",
      "value": 0.00016482500000000036,
      "lo": 0.00016482500000000036,
      "hi": 0.00016482500000000036,
      "n": 360,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "capture=scan",
      "target": "decision:gpt-6-luna",
      "value": 0.00016482500000000036,
      "lo": 0.00016482500000000036,
      "hi": 0.00016482500000000036,
      "n": 360,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "capture=scan",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.010558056888693507,
      "n": 360,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "capture=scan",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.010558056888693507,
      "n": 360,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "capture=scan",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.010558056888693507,
      "n": 360,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "capture=screenshot",
      "target": "decision:gpt-6-luna",
      "value": 342,
      "lo": 342,
      "hi": 342,
      "n": 15,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "capture=screenshot",
      "target": "decision:gpt-6-luna",
      "value": 490.2,
      "lo": 490.2,
      "hi": 490.2,
      "n": 15,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "capture=screenshot",
      "target": "decision:gpt-6-luna",
      "value": 509.24,
      "lo": 509.24,
      "hi": 509.24,
      "n": 15,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "capture=screenshot",
      "target": "decision:gpt-6-luna",
      "value": 0.000154,
      "lo": 0.000154,
      "hi": 0.000154,
      "n": 15,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "capture=screenshot",
      "target": "decision:gpt-6-luna",
      "value": 0.000154,
      "lo": 0.000154,
      "hi": 0.000154,
      "n": 15,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "capture=screenshot",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "capture=screenshot",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "capture=screenshot",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "class=ai_full",
      "target": "decision:gpt-6-luna",
      "value": 392.5,
      "lo": 392.5,
      "hi": 392.5,
      "n": 60,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "class=ai_full",
      "target": "decision:gpt-6-luna",
      "value": 769.75,
      "lo": 769.75,
      "hi": 769.75,
      "n": 60,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "class=ai_full",
      "target": "decision:gpt-6-luna",
      "value": 903.0399999999995,
      "lo": 903.0399999999995,
      "hi": 903.0399999999995,
      "n": 60,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "class=ai_full",
      "target": "decision:gpt-6-luna",
      "value": 0.0001762999999999999,
      "lo": 0.0001762999999999999,
      "hi": 0.0001762999999999999,
      "n": 60,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "class=ai_full",
      "target": "decision:gpt-6-luna",
      "value": 0.0001762999999999999,
      "lo": 0.0001762999999999999,
      "hi": 0.0001762999999999999,
      "n": 60,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "class=ai_full",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.06017185214208986,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "class=ai_full",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.06017185214208986,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "class=ai_full",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.06017185214208986,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "class=ai_inpaint",
      "target": "decision:gpt-6-luna",
      "value": 348.5,
      "lo": 348.5,
      "hi": 348.5,
      "n": 30,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "class=ai_inpaint",
      "target": "decision:gpt-6-luna",
      "value": 446.59999999999997,
      "lo": 446.59999999999997,
      "hi": 446.59999999999997,
      "n": 30,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "class=ai_inpaint",
      "target": "decision:gpt-6-luna",
      "value": 452,
      "lo": 452,
      "hi": 452,
      "n": 30,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "class=ai_inpaint",
      "target": "decision:gpt-6-luna",
      "value": 0.0001798000000000001,
      "lo": 0.0001798000000000001,
      "hi": 0.0001798000000000001,
      "n": 30,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "class=ai_inpaint",
      "target": "decision:gpt-6-luna",
      "value": 0.0001798000000000001,
      "lo": 0.0001798000000000001,
      "hi": 0.0001798000000000001,
      "n": 30,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "class=ai_inpaint",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "class=ai_inpaint",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "class=ai_inpaint",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "class=duplicate",
      "target": "decision:gpt-6-luna",
      "value": 336,
      "lo": 336,
      "hi": 336,
      "n": 90,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "class=duplicate",
      "target": "decision:gpt-6-luna",
      "value": 482.75,
      "lo": 482.75,
      "hi": 482.75,
      "n": 90,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "class=duplicate",
      "target": "decision:gpt-6-luna",
      "value": 687.5999999999999,
      "lo": 687.5999999999999,
      "hi": 687.5999999999999,
      "n": 90,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "class=duplicate",
      "target": "decision:gpt-6-luna",
      "value": 0.00015826666666666673,
      "lo": 0.00015826666666666673,
      "hi": 0.00015826666666666673,
      "n": 90,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "class=duplicate",
      "target": "decision:gpt-6-luna",
      "value": 0.00015826666666666673,
      "lo": 0.00015826666666666673,
      "hi": 0.00015826666666666673,
      "n": 90,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "class=duplicate",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.04093562556432678,
      "n": 90,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "class=duplicate",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.04093562556432678,
      "n": 90,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "class=duplicate",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.04093562556432678,
      "n": 90,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "class=genuine",
      "target": "decision:gpt-6-luna",
      "value": 345,
      "lo": 345,
      "hi": 345,
      "n": 210,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "class=genuine",
      "target": "decision:gpt-6-luna",
      "value": 532.4999999999998,
      "lo": 532.4999999999998,
      "hi": 532.4999999999998,
      "n": 210,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "class=genuine",
      "target": "decision:gpt-6-luna",
      "value": 806.3799999999999,
      "lo": 806.3799999999999,
      "hi": 806.3799999999999,
      "n": 210,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "class=genuine",
      "target": "decision:gpt-6-luna",
      "value": 0.000168557142857143,
      "lo": 0.000168557142857143,
      "hi": 0.000168557142857143,
      "n": 210,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "class=genuine",
      "target": "decision:gpt-6-luna",
      "value": 0.000168557142857143,
      "lo": 0.000168557142857143,
      "hi": 0.000168557142857143,
      "n": 210,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "class=genuine",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.017964050759283236,
      "n": 210,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "class=genuine",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.017964050759283236,
      "n": 210,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "class=genuine",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.017964050759283236,
      "n": 210,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "class=hard_negative",
      "target": "decision:gpt-6-luna",
      "value": 374,
      "lo": 374,
      "hi": 374,
      "n": 33,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "class=hard_negative",
      "target": "decision:gpt-6-luna",
      "value": 605.1999999999996,
      "lo": 605.1999999999996,
      "hi": 605.1999999999996,
      "n": 33,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "class=hard_negative",
      "target": "decision:gpt-6-luna",
      "value": 6312.479999999998,
      "lo": 6312.479999999998,
      "hi": 6312.479999999998,
      "n": 33,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "class=hard_negative",
      "target": "decision:gpt-6-luna",
      "value": 0.0001742727272727273,
      "lo": 0.0001742727272727273,
      "hi": 0.0001742727272727273,
      "n": 33,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "class=hard_negative",
      "target": "decision:gpt-6-luna",
      "value": 0.0001742727272727273,
      "lo": 0.0001742727272727273,
      "hi": 0.0001742727272727273,
      "n": 33,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "class=hard_negative",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.10426999754245206,
      "n": 33,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "class=hard_negative",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.10426999754245206,
      "n": 33,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "class=hard_negative",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.10426999754245206,
      "n": 33,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "class=manual_edit",
      "target": "decision:gpt-6-luna",
      "value": 328,
      "lo": 328,
      "hi": 328,
      "n": 30,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "class=manual_edit",
      "target": "decision:gpt-6-luna",
      "value": 580.0999999999996,
      "lo": 580.0999999999996,
      "hi": 580.0999999999996,
      "n": 30,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "class=manual_edit",
      "target": "decision:gpt-6-luna",
      "value": 658.91,
      "lo": 658.91,
      "hi": 658.91,
      "n": 30,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "class=manual_edit",
      "target": "decision:gpt-6-luna",
      "value": 0.0001597,
      "lo": 0.0001597,
      "hi": 0.0001597,
      "n": 30,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "class=manual_edit",
      "target": "decision:gpt-6-luna",
      "value": 0.0001597,
      "lo": 0.0001597,
      "hi": 0.0001597,
      "n": 30,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "class=manual_edit",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "class=manual_edit",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "class=manual_edit",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "country=JP",
      "target": "decision:gpt-6-luna",
      "value": 433.5,
      "lo": 433.5,
      "hi": 433.5,
      "n": 12,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "country=JP",
      "target": "decision:gpt-6-luna",
      "value": 532.85,
      "lo": 532.85,
      "hi": 532.85,
      "n": 12,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "country=JP",
      "target": "decision:gpt-6-luna",
      "value": 538.57,
      "lo": 538.57,
      "hi": 538.57,
      "n": 12,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "country=JP",
      "target": "decision:gpt-6-luna",
      "value": 0.00021300000000000005,
      "lo": 0.00021300000000000005,
      "hi": 0.00021300000000000005,
      "n": 12,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "country=JP",
      "target": "decision:gpt-6-luna",
      "value": 0.00021300000000000005,
      "lo": 0.00021300000000000005,
      "hi": 0.00021300000000000005,
      "n": 12,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "country=JP",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "country=JP",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "country=JP",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "country=MY",
      "target": "decision:gpt-6-luna",
      "value": 343,
      "lo": 343,
      "hi": 343,
      "n": 369,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "country=MY",
      "target": "decision:gpt-6-luna",
      "value": 506.7999999999997,
      "lo": 506.7999999999997,
      "hi": 506.7999999999997,
      "n": 369,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "country=MY",
      "target": "decision:gpt-6-luna",
      "value": 741.9199999999998,
      "lo": 741.9199999999998,
      "hi": 741.9199999999998,
      "n": 369,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "country=MY",
      "target": "decision:gpt-6-luna",
      "value": 0.000165756097560976,
      "lo": 0.000165756097560976,
      "hi": 0.000165756097560976,
      "n": 369,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "country=MY",
      "target": "decision:gpt-6-luna",
      "value": 0.000165756097560976,
      "lo": 0.000165756097560976,
      "hi": 0.000165756097560976,
      "n": 369,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "country=MY",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.010303196519090837,
      "n": 369,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "country=MY",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.010303196519090837,
      "n": 369,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "country=MY",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.010303196519090837,
      "n": 369,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "currency=JPY",
      "target": "decision:gpt-6-luna",
      "value": 433.5,
      "lo": 433.5,
      "hi": 433.5,
      "n": 12,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "currency=JPY",
      "target": "decision:gpt-6-luna",
      "value": 532.85,
      "lo": 532.85,
      "hi": 532.85,
      "n": 12,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "currency=JPY",
      "target": "decision:gpt-6-luna",
      "value": 538.57,
      "lo": 538.57,
      "hi": 538.57,
      "n": 12,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "currency=JPY",
      "target": "decision:gpt-6-luna",
      "value": 0.00021300000000000005,
      "lo": 0.00021300000000000005,
      "hi": 0.00021300000000000005,
      "n": 12,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "currency=JPY",
      "target": "decision:gpt-6-luna",
      "value": 0.00021300000000000005,
      "lo": 0.00021300000000000005,
      "hi": 0.00021300000000000005,
      "n": 12,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "currency=JPY",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "currency=JPY",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "currency=JPY",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "currency=MYR",
      "target": "decision:gpt-6-luna",
      "value": 343,
      "lo": 343,
      "hi": 343,
      "n": 369,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "currency=MYR",
      "target": "decision:gpt-6-luna",
      "value": 506.7999999999997,
      "lo": 506.7999999999997,
      "hi": 506.7999999999997,
      "n": 369,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "currency=MYR",
      "target": "decision:gpt-6-luna",
      "value": 741.9199999999998,
      "lo": 741.9199999999998,
      "hi": 741.9199999999998,
      "n": 369,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "currency=MYR",
      "target": "decision:gpt-6-luna",
      "value": 0.000165756097560976,
      "lo": 0.000165756097560976,
      "hi": 0.000165756097560976,
      "n": 369,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "currency=MYR",
      "target": "decision:gpt-6-luna",
      "value": 0.000165756097560976,
      "lo": 0.000165756097560976,
      "hi": 0.000165756097560976,
      "n": 369,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "currency=MYR",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.010303196519090837,
      "n": 369,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "currency=MYR",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.010303196519090837,
      "n": 369,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "currency=MYR",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.010303196519090837,
      "n": 369,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "dateEra=old",
      "target": "decision:gpt-6-luna",
      "value": 344,
      "lo": 344,
      "hi": 344,
      "n": 384,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "dateEra=old",
      "target": "decision:gpt-6-luna",
      "value": 534.5999999999995,
      "lo": 534.5999999999995,
      "hi": 534.5999999999995,
      "n": 384,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "dateEra=old",
      "target": "decision:gpt-6-luna",
      "value": 771.5500000000002,
      "lo": 771.5500000000002,
      "hi": 771.5500000000002,
      "n": 384,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "dateEra=old",
      "target": "decision:gpt-6-luna",
      "value": 0.00016628125000000044,
      "lo": 0.00016628125000000044,
      "hi": 0.00016628125000000044,
      "n": 384,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "dateEra=old",
      "target": "decision:gpt-6-luna",
      "value": 0.00016628125000000044,
      "lo": 0.00016628125000000044,
      "hi": 0.00016628125000000044,
      "n": 384,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "dateEra=old",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 8.673617379884035e-19,
      "hi": 0.009904714241677031,
      "n": 384,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "dateEra=old",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 8.673617379884035e-19,
      "hi": 0.009904714241677031,
      "n": 384,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "dateEra=old",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 8.673617379884035e-19,
      "hi": 0.009904714241677031,
      "n": 384,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "dateEra=recent",
      "target": "decision:gpt-6-luna",
      "value": 375,
      "lo": 375,
      "hi": 375,
      "n": 69,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "dateEra=recent",
      "target": "decision:gpt-6-luna",
      "value": 820.5999999999999,
      "lo": 820.5999999999999,
      "hi": 820.5999999999999,
      "n": 69,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "dateEra=recent",
      "target": "decision:gpt-6-luna",
      "value": 1019.9999999999993,
      "lo": 1019.9999999999993,
      "hi": 1019.9999999999993,
      "n": 69,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "dateEra=recent",
      "target": "decision:gpt-6-luna",
      "value": 0.00017830434782608703,
      "lo": 0.00017830434782608703,
      "hi": 0.00017830434782608703,
      "n": 69,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "dateEra=recent",
      "target": "decision:gpt-6-luna",
      "value": 0.00017830434782608703,
      "lo": 0.00017830434782608703,
      "hi": 0.00017830434782608703,
      "n": 69,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "dateEra=recent",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 3.469446951953614e-18,
      "hi": 0.05273725818905173,
      "n": 69,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "dateEra=recent",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 3.469446951953614e-18,
      "hi": 0.05273725818905173,
      "n": 69,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "dateEra=recent",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 3.469446951953614e-18,
      "hi": 0.05273725818905173,
      "n": 69,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "docType=receipt",
      "target": "decision:gpt-6-luna",
      "value": 349,
      "lo": 349,
      "hi": 349,
      "n": 453,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "docType=receipt",
      "target": "decision:gpt-6-luna",
      "value": 565.7999999999995,
      "lo": 565.7999999999995,
      "hi": 565.7999999999995,
      "n": 453,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "docType=receipt",
      "target": "decision:gpt-6-luna",
      "value": 829.48,
      "lo": 829.48,
      "hi": 829.48,
      "n": 453,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "docType=receipt",
      "target": "decision:gpt-6-luna",
      "value": 0.0001681125827814572,
      "lo": 0.0001681125827814572,
      "hi": 0.0001681125827814572,
      "n": 453,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "docType=receipt",
      "target": "decision:gpt-6-luna",
      "value": 0.0001681125827814572,
      "lo": 0.0001681125827814572,
      "hi": 0.0001681125827814572,
      "n": 453,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "docType=receipt",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.008408735123582253,
      "n": 453,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "docType=receipt",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.008408735123582253,
      "n": 453,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "docType=receipt",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.008408735123582253,
      "n": 453,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "editedField=date",
      "target": "decision:gpt-6-luna",
      "value": 355.5,
      "lo": 355.5,
      "hi": 355.5,
      "n": 12,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "editedField=date",
      "target": "decision:gpt-6-luna",
      "value": 452.2,
      "lo": 452.2,
      "hi": 452.2,
      "n": 12,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "editedField=date",
      "target": "decision:gpt-6-luna",
      "value": 468.04,
      "lo": 468.04,
      "hi": 468.04,
      "n": 12,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "editedField=date",
      "target": "decision:gpt-6-luna",
      "value": 0.00017774999999999998,
      "lo": 0.00017774999999999998,
      "hi": 0.00017774999999999998,
      "n": 12,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "editedField=date",
      "target": "decision:gpt-6-luna",
      "value": 0.00017774999999999998,
      "lo": 0.00017774999999999998,
      "hi": 0.00017774999999999998,
      "n": 12,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "editedField=date",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "editedField=date",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "editedField=date",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "editedField=items+total",
      "target": "decision:gpt-6-luna",
      "value": 364,
      "lo": 364,
      "hi": 364,
      "n": 12,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "editedField=items+total",
      "target": "decision:gpt-6-luna",
      "value": 463.25,
      "lo": 463.25,
      "hi": 463.25,
      "n": 12,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "editedField=items+total",
      "target": "decision:gpt-6-luna",
      "value": 474.25,
      "lo": 474.25,
      "hi": 474.25,
      "n": 12,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "editedField=items+total",
      "target": "decision:gpt-6-luna",
      "value": 0.00017675000000000004,
      "lo": 0.00017675000000000004,
      "hi": 0.00017675000000000004,
      "n": 12,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "editedField=items+total",
      "target": "decision:gpt-6-luna",
      "value": 0.00017675000000000004,
      "lo": 0.00017675000000000004,
      "hi": 0.00017675000000000004,
      "n": 12,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "editedField=items+total",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "editedField=items+total",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "editedField=items+total",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "editedField=total",
      "target": "decision:gpt-6-luna",
      "value": 340,
      "lo": 340,
      "hi": 340,
      "n": 27,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "editedField=total",
      "target": "decision:gpt-6-luna",
      "value": 601.3999999999999,
      "lo": 601.3999999999999,
      "hi": 601.3999999999999,
      "n": 27,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "editedField=total",
      "target": "decision:gpt-6-luna",
      "value": 659.54,
      "lo": 659.54,
      "hi": 659.54,
      "n": 27,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "editedField=total",
      "target": "decision:gpt-6-luna",
      "value": 0.00016099999999999998,
      "lo": 0.00016099999999999998,
      "hi": 0.00016099999999999998,
      "n": 27,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "editedField=total",
      "target": "decision:gpt-6-luna",
      "value": 0.00016099999999999998,
      "lo": 0.00016099999999999998,
      "hi": 0.00016099999999999998,
      "n": 27,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "editedField=total",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.12455502974186705,
      "n": 27,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "editedField=total",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.12455502974186705,
      "n": 27,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "editedField=total",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.12455502974186705,
      "n": 27,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "lang=en",
      "target": "decision:gpt-6-luna",
      "value": 344,
      "lo": 344,
      "hi": 344,
      "n": 387,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "lang=en",
      "target": "decision:gpt-6-luna",
      "value": 512.8,
      "lo": 512.8,
      "hi": 512.8,
      "n": 387,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "lang=en",
      "target": "decision:gpt-6-luna",
      "value": 765.9399999999997,
      "lo": 765.9399999999997,
      "hi": 765.9399999999997,
      "n": 387,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "lang=en",
      "target": "decision:gpt-6-luna",
      "value": 0.00016654263565891517,
      "lo": 0.00016654263565891517,
      "hi": 0.00016654263565891517,
      "n": 387,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "lang=en",
      "target": "decision:gpt-6-luna",
      "value": 0.00016654263565891517,
      "lo": 0.00016654263565891517,
      "hi": 0.00016654263565891517,
      "n": 387,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "lang=en",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.009828688164979105,
      "n": 387,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "lang=en",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.009828688164979105,
      "n": 387,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "lang=en",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.009828688164979105,
      "n": 387,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "lang=ja",
      "target": "decision:gpt-6-luna",
      "value": 433.5,
      "lo": 433.5,
      "hi": 433.5,
      "n": 12,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "lang=ja",
      "target": "decision:gpt-6-luna",
      "value": 532.85,
      "lo": 532.85,
      "hi": 532.85,
      "n": 12,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "lang=ja",
      "target": "decision:gpt-6-luna",
      "value": 538.57,
      "lo": 538.57,
      "hi": 538.57,
      "n": 12,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "lang=ja",
      "target": "decision:gpt-6-luna",
      "value": 0.00021300000000000005,
      "lo": 0.00021300000000000005,
      "hi": 0.00021300000000000005,
      "n": 12,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "lang=ja",
      "target": "decision:gpt-6-luna",
      "value": 0.00021300000000000005,
      "lo": 0.00021300000000000005,
      "hi": 0.00021300000000000005,
      "n": 12,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "lang=ja",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "lang=ja",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "lang=ja",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "lang=zh",
      "target": "decision:gpt-6-luna",
      "value": 345,
      "lo": 345,
      "hi": 345,
      "n": 18,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "lang=zh",
      "target": "decision:gpt-6-luna",
      "value": 600.0499999999996,
      "lo": 600.0499999999996,
      "hi": 600.0499999999996,
      "n": 18,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "lang=zh",
      "target": "decision:gpt-6-luna",
      "value": 795.2099999999996,
      "lo": 795.2099999999996,
      "hi": 795.2099999999996,
      "n": 18,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "lang=zh",
      "target": "decision:gpt-6-luna",
      "value": 0.00016900000000000002,
      "lo": 0.00016900000000000002,
      "hi": 0.00016900000000000002,
      "n": 18,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "lang=zh",
      "target": "decision:gpt-6-luna",
      "value": 0.00016900000000000002,
      "lo": 0.00016900000000000002,
      "hi": 0.00016900000000000002,
      "n": 18,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "lang=zh",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.17587922364665765,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "lang=zh",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.17587922364665765,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "lang=zh",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.17587922364665765,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "prompt=explicit_amounts",
      "target": "decision:gpt-6-luna",
      "value": 396.5,
      "lo": 396.5,
      "hi": 396.5,
      "n": 24,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "prompt=explicit_amounts",
      "target": "decision:gpt-6-luna",
      "value": 776.1999999999999,
      "lo": 776.1999999999999,
      "hi": 776.1999999999999,
      "n": 24,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "prompt=explicit_amounts",
      "target": "decision:gpt-6-luna",
      "value": 830.1999999999999,
      "lo": 830.1999999999999,
      "hi": 830.1999999999999,
      "n": 24,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "prompt=explicit_amounts",
      "target": "decision:gpt-6-luna",
      "value": 0.00017774999999999993,
      "lo": 0.00017774999999999993,
      "hi": 0.00017774999999999993,
      "n": 24,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "prompt=explicit_amounts",
      "target": "decision:gpt-6-luna",
      "value": 0.00017774999999999993,
      "lo": 0.00017774999999999993,
      "hi": 0.00017774999999999993,
      "n": 24,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "prompt=explicit_amounts",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 1.3877787807814457e-17,
      "hi": 0.13797620467498017,
      "n": 24,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "prompt=explicit_amounts",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 1.3877787807814457e-17,
      "hi": 0.13797620467498017,
      "n": 24,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "prompt=explicit_amounts",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 1.3877787807814457e-17,
      "hi": 0.13797620467498017,
      "n": 24,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "prompt=free",
      "target": "decision:gpt-6-luna",
      "value": 414,
      "lo": 414,
      "hi": 414,
      "n": 15,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "prompt=free",
      "target": "decision:gpt-6-luna",
      "value": 834.6999999999998,
      "lo": 834.6999999999998,
      "hi": 834.6999999999998,
      "n": 15,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "prompt=free",
      "target": "decision:gpt-6-luna",
      "value": 957.3399999999999,
      "lo": 957.3399999999999,
      "hi": 957.3399999999999,
      "n": 15,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "prompt=free",
      "target": "decision:gpt-6-luna",
      "value": 0.00017919999999999996,
      "lo": 0.00017919999999999996,
      "hi": 0.00017919999999999996,
      "n": 15,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "prompt=free",
      "target": "decision:gpt-6-luna",
      "value": 0.00017919999999999996,
      "lo": 0.00017919999999999996,
      "hi": 0.00017919999999999996,
      "n": 15,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "prompt=free",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "prompt=free",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "prompt=free",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "prompt=target_total",
      "target": "decision:gpt-6-luna",
      "value": 374,
      "lo": 374,
      "hi": 374,
      "n": 21,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "prompt=target_total",
      "target": "decision:gpt-6-luna",
      "value": 537,
      "lo": 537,
      "hi": 537,
      "n": 21,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "prompt=target_total",
      "target": "decision:gpt-6-luna",
      "value": 585.8000000000001,
      "lo": 585.8000000000001,
      "hi": 585.8000000000001,
      "n": 21,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "prompt=target_total",
      "target": "decision:gpt-6-luna",
      "value": 0.0001725714285714285,
      "lo": 0.0001725714285714285,
      "hi": 0.0001725714285714285,
      "n": 21,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "prompt=target_total",
      "target": "decision:gpt-6-luna",
      "value": 0.0001725714285714285,
      "lo": 0.0001725714285714285,
      "hi": 0.0001725714285714285,
      "n": 21,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "prompt=target_total",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.154639018924847,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "prompt=target_total",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.154639018924847,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "prompt=target_total",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.154639018924847,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "provenance=intact",
      "target": "decision:gpt-6-luna",
      "value": 375,
      "lo": 375,
      "hi": 375,
      "n": 60,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "provenance=intact",
      "target": "decision:gpt-6-luna",
      "value": 661.6999999999998,
      "lo": 661.6999999999998,
      "hi": 661.6999999999998,
      "n": 60,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "provenance=intact",
      "target": "decision:gpt-6-luna",
      "value": 867.6399999999993,
      "lo": 867.6399999999993,
      "hi": 867.6399999999993,
      "n": 60,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "provenance=intact",
      "target": "decision:gpt-6-luna",
      "value": 0.00017940000000000002,
      "lo": 0.00017940000000000002,
      "hi": 0.00017940000000000002,
      "n": 60,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "provenance=intact",
      "target": "decision:gpt-6-luna",
      "value": 0.00017940000000000002,
      "lo": 0.00017940000000000002,
      "hi": 0.00017940000000000002,
      "n": 60,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "provenance=intact",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.06017185214208986,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "provenance=intact",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.06017185214208986,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "provenance=intact",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.06017185214208986,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "provenance=none",
      "target": "decision:gpt-6-luna",
      "value": 346,
      "lo": 346,
      "hi": 346,
      "n": 333,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "provenance=none",
      "target": "decision:gpt-6-luna",
      "value": 523.3999999999999,
      "lo": 523.3999999999999,
      "hi": 523.3999999999999,
      "n": 333,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "provenance=none",
      "target": "decision:gpt-6-luna",
      "value": 822.2800000000002,
      "lo": 822.2800000000002,
      "hi": 822.2800000000002,
      "n": 333,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "provenance=none",
      "target": "decision:gpt-6-luna",
      "value": 0.0001663423423423424,
      "lo": 0.0001663423423423424,
      "hi": 0.0001663423423423424,
      "n": 333,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "provenance=none",
      "target": "decision:gpt-6-luna",
      "value": 0.0001663423423423424,
      "lo": 0.0001663423423423424,
      "hi": 0.0001663423423423424,
      "n": 333,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "provenance=none",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.011404352760326315,
      "n": 333,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "provenance=none",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.011404352760326315,
      "n": 333,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "provenance=none",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.011404352760326315,
      "n": 333,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "provenance=stripped",
      "target": "decision:gpt-6-luna",
      "value": 352,
      "lo": 352,
      "hi": 352,
      "n": 60,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "provenance=stripped",
      "target": "decision:gpt-6-luna",
      "value": 645.05,
      "lo": 645.05,
      "hi": 645.05,
      "n": 60,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "provenance=stripped",
      "target": "decision:gpt-6-luna",
      "value": 799.7499999999998,
      "lo": 799.7499999999998,
      "hi": 799.7499999999998,
      "n": 60,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "provenance=stripped",
      "target": "decision:gpt-6-luna",
      "value": 0.00016665000000000004,
      "lo": 0.00016665000000000004,
      "hi": 0.00016665000000000004,
      "n": 60,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "provenance=stripped",
      "target": "decision:gpt-6-luna",
      "value": 0.00016665000000000004,
      "lo": 0.00016665000000000004,
      "hi": 0.00016665000000000004,
      "n": 60,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "provenance=stripped",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.06017185214208986,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "provenance=stripped",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.06017185214208986,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "provenance=stripped",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.06017185214208986,
      "n": 60,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "source=commons",
      "target": "decision:gpt-6-luna",
      "value": 374,
      "lo": 374,
      "hi": 374,
      "n": 39,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "source=commons",
      "target": "decision:gpt-6-luna",
      "value": 810.1,
      "lo": 810.1,
      "hi": 810.1,
      "n": 39,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "source=commons",
      "target": "decision:gpt-6-luna",
      "value": 989.5799999999994,
      "lo": 989.5799999999994,
      "hi": 989.5799999999994,
      "n": 39,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "source=commons",
      "target": "decision:gpt-6-luna",
      "value": 0.0001780769230769231,
      "lo": 0.0001780769230769231,
      "hi": 0.0001780769230769231,
      "n": 39,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "source=commons",
      "target": "decision:gpt-6-luna",
      "value": 0.0001780769230769231,
      "lo": 0.0001780769230769231,
      "hi": 0.0001780769230769231,
      "n": 39,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "source=commons",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.08966685370757144,
      "n": 39,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "source=commons",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.08966685370757144,
      "n": 39,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "source=commons",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.08966685370757144,
      "n": 39,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "source=edited:gemini-nano-banana-2.1",
      "target": "decision:gpt-6-luna",
      "value": 342,
      "lo": 342,
      "hi": 342,
      "n": 12,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "source=edited:gemini-nano-banana-2.1",
      "target": "decision:gpt-6-luna",
      "value": 359,
      "lo": 359,
      "hi": 359,
      "n": 12,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "source=edited:gemini-nano-banana-2.1",
      "target": "decision:gpt-6-luna",
      "value": 359,
      "lo": 359,
      "hi": 359,
      "n": 12,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "source=edited:gemini-nano-banana-2.1",
      "target": "decision:gpt-6-luna",
      "value": 0.0001595,
      "lo": 0.0001595,
      "hi": 0.0001595,
      "n": 12,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "source=edited:gemini-nano-banana-2.1",
      "target": "decision:gpt-6-luna",
      "value": 0.0001595,
      "lo": 0.0001595,
      "hi": 0.0001595,
      "n": 12,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "source=edited:gemini-nano-banana-2.1",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "source=edited:gemini-nano-banana-2.1",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "source=edited:gemini-nano-banana-2.1",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "source=edited:gpt-image-2.5-sunburst",
      "target": "decision:gpt-6-luna",
      "value": 409.5,
      "lo": 409.5,
      "hi": 409.5,
      "n": 12,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "source=edited:gpt-image-2.5-sunburst",
      "target": "decision:gpt-6-luna",
      "value": 452,
      "lo": 452,
      "hi": 452,
      "n": 12,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "source=edited:gpt-image-2.5-sunburst",
      "target": "decision:gpt-6-luna",
      "value": 452,
      "lo": 452,
      "hi": 452,
      "n": 12,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "source=edited:gpt-image-2.5-sunburst",
      "target": "decision:gpt-6-luna",
      "value": 0.0002105,
      "lo": 0.0002105,
      "hi": 0.0002105,
      "n": 12,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "source=edited:gpt-image-2.5-sunburst",
      "target": "decision:gpt-6-luna",
      "value": 0.0002105,
      "lo": 0.0002105,
      "hi": 0.0002105,
      "n": 12,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "source=edited:gpt-image-2.5-sunburst",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "source=edited:gpt-image-2.5-sunburst",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "source=edited:gpt-image-2.5-sunburst",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 2.7755575615628914e-17,
      "hi": 0.24249400665524085,
      "n": 12,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "source=edited:sharp-patch",
      "target": "decision:gpt-6-luna",
      "value": 328,
      "lo": 328,
      "hi": 328,
      "n": 30,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "source=edited:sharp-patch",
      "target": "decision:gpt-6-luna",
      "value": 580.0999999999996,
      "lo": 580.0999999999996,
      "hi": 580.0999999999996,
      "n": 30,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "source=edited:sharp-patch",
      "target": "decision:gpt-6-luna",
      "value": 658.91,
      "lo": 658.91,
      "hi": 658.91,
      "n": 30,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "source=edited:sharp-patch",
      "target": "decision:gpt-6-luna",
      "value": 0.0001597,
      "lo": 0.0001597,
      "hi": 0.0001597,
      "n": 30,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "source=edited:sharp-patch",
      "target": "decision:gpt-6-luna",
      "value": 0.0001597,
      "lo": 0.0001597,
      "hi": 0.0001597,
      "n": 30,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "source=edited:sharp-patch",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "source=edited:sharp-patch",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "source=edited:sharp-patch",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.11351339317396876,
      "n": 30,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "source=generated:flux-3-image",
      "target": "decision:gpt-6-luna",
      "value": 350,
      "lo": 350,
      "hi": 350,
      "n": 18,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "source=generated:flux-3-image",
      "target": "decision:gpt-6-luna",
      "value": 780.2499999999999,
      "lo": 780.2499999999999,
      "hi": 780.2499999999999,
      "n": 18,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "source=generated:flux-3-image",
      "target": "decision:gpt-6-luna",
      "value": 831.2499999999999,
      "lo": 831.2499999999999,
      "hi": 831.2499999999999,
      "n": 18,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "source=generated:flux-3-image",
      "target": "decision:gpt-6-luna",
      "value": 0.00015599999999999994,
      "lo": 0.00015599999999999994,
      "hi": 0.00015599999999999994,
      "n": 18,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "source=generated:flux-3-image",
      "target": "decision:gpt-6-luna",
      "value": 0.00015599999999999994,
      "lo": 0.00015599999999999994,
      "hi": 0.00015599999999999994,
      "n": 18,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "source=generated:flux-3-image",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.17587922364665765,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "source=generated:flux-3-image",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.17587922364665765,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "source=generated:flux-3-image",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.17587922364665765,
      "n": 18,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "source=generated:gemini-nano-banana-2.1",
      "target": "decision:gpt-6-luna",
      "value": 376,
      "lo": 376,
      "hi": 376,
      "n": 21,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "source=generated:gemini-nano-banana-2.1",
      "target": "decision:gpt-6-luna",
      "value": 537,
      "lo": 537,
      "hi": 537,
      "n": 21,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "source=generated:gemini-nano-banana-2.1",
      "target": "decision:gpt-6-luna",
      "value": 734.6000000000001,
      "lo": 734.6000000000001,
      "hi": 734.6000000000001,
      "n": 21,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "source=generated:gemini-nano-banana-2.1",
      "target": "decision:gpt-6-luna",
      "value": 0.00016199999999999995,
      "lo": 0.00016199999999999995,
      "hi": 0.00016199999999999995,
      "n": 21,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "source=generated:gemini-nano-banana-2.1",
      "target": "decision:gpt-6-luna",
      "value": 0.00016199999999999995,
      "lo": 0.00016199999999999995,
      "hi": 0.00016199999999999995,
      "n": 21,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "source=generated:gemini-nano-banana-2.1",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.154639018924847,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "source=generated:gemini-nano-banana-2.1",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.154639018924847,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "source=generated:gemini-nano-banana-2.1",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.154639018924847,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "source=generated:gpt-image-2.5-flare",
      "target": "decision:gpt-6-luna",
      "value": 408,
      "lo": 408,
      "hi": 408,
      "n": 21,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "source=generated:gpt-image-2.5-flare",
      "target": "decision:gpt-6-luna",
      "value": 598,
      "lo": 598,
      "hi": 598,
      "n": 21,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "source=generated:gpt-image-2.5-flare",
      "target": "decision:gpt-6-luna",
      "value": 910.0000000000002,
      "lo": 910.0000000000002,
      "hi": 910.0000000000002,
      "n": 21,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "source=generated:gpt-image-2.5-flare",
      "target": "decision:gpt-6-luna",
      "value": 0.00020799999999999993,
      "lo": 0.00020799999999999993,
      "hi": 0.00020799999999999993,
      "n": 21,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "source=generated:gpt-image-2.5-flare",
      "target": "decision:gpt-6-luna",
      "value": 0.00020799999999999993,
      "lo": 0.00020799999999999993,
      "hi": 0.00020799999999999993,
      "n": 21,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "source=generated:gpt-image-2.5-flare",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.154639018924847,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "source=generated:gpt-image-2.5-flare",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.154639018924847,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "source=generated:gpt-image-2.5-flare",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.154639018924847,
      "n": 21,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "source=sroie",
      "target": "decision:gpt-6-luna",
      "value": 342,
      "lo": 342,
      "hi": 342,
      "n": 294,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "source=sroie",
      "target": "decision:gpt-6-luna",
      "value": 500.34999999999997,
      "lo": 500.34999999999997,
      "hi": 500.34999999999997,
      "n": 294,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "source=sroie",
      "target": "decision:gpt-6-luna",
      "value": 764.8899999999999,
      "lo": 764.8899999999999,
      "hi": 764.8899999999999,
      "n": 294,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "source=sroie",
      "target": "decision:gpt-6-luna",
      "value": 0.0001647857142857145,
      "lo": 0.0001647857142857145,
      "hi": 0.0001647857142857145,
      "n": 294,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "source=sroie",
      "target": "decision:gpt-6-luna",
      "value": 0.0001647857142857145,
      "lo": 0.0001647857142857145,
      "hi": 0.0001647857142857145,
      "n": 294,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "source=sroie",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.012897663192708011,
      "n": 294,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "source=sroie",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.012897663192708011,
      "n": 294,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "source=sroie",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.012897663192708011,
      "n": 294,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "transform=crop",
      "target": "decision:gpt-6-luna",
      "value": 334,
      "lo": 334,
      "hi": 334,
      "n": 15,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "transform=crop",
      "target": "decision:gpt-6-luna",
      "value": 406.49999999999994,
      "lo": 406.49999999999994,
      "hi": 406.49999999999994,
      "n": 15,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "transform=crop",
      "target": "decision:gpt-6-luna",
      "value": 426.09999999999997,
      "lo": 426.09999999999997,
      "hi": 426.09999999999997,
      "n": 15,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "transform=crop",
      "target": "decision:gpt-6-luna",
      "value": 0.00015000000000000001,
      "lo": 0.00015000000000000001,
      "hi": 0.00015000000000000001,
      "n": 15,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "transform=crop",
      "target": "decision:gpt-6-luna",
      "value": 0.00015000000000000001,
      "lo": 0.00015000000000000001,
      "hi": 0.00015000000000000001,
      "n": 15,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "transform=crop",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "transform=crop",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "transform=crop",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "transform=jpeg",
      "target": "decision:gpt-6-luna",
      "value": 320,
      "lo": 320,
      "hi": 320,
      "n": 15,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "transform=jpeg",
      "target": "decision:gpt-6-luna",
      "value": 378.19999999999993,
      "lo": 378.19999999999993,
      "hi": 378.19999999999993,
      "n": 15,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "transform=jpeg",
      "target": "decision:gpt-6-luna",
      "value": 408.43999999999994,
      "lo": 408.43999999999994,
      "hi": 408.43999999999994,
      "n": 15,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "transform=jpeg",
      "target": "decision:gpt-6-luna",
      "value": 0.00017760000000000003,
      "lo": 0.00017760000000000003,
      "hi": 0.00017760000000000003,
      "n": 15,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "transform=jpeg",
      "target": "decision:gpt-6-luna",
      "value": 0.00017760000000000003,
      "lo": 0.00017760000000000003,
      "hi": 0.00017760000000000003,
      "n": 15,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "transform=jpeg",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "transform=jpeg",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "transform=jpeg",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "transform=rephoto",
      "target": "decision:gpt-6-luna",
      "value": 364,
      "lo": 364,
      "hi": 364,
      "n": 15,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "transform=rephoto",
      "target": "decision:gpt-6-luna",
      "value": 496.0999999999998,
      "lo": 496.0999999999998,
      "hi": 496.0999999999998,
      "n": 15,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "transform=rephoto",
      "target": "decision:gpt-6-luna",
      "value": 617.6199999999999,
      "lo": 617.6199999999999,
      "hi": 617.6199999999999,
      "n": 15,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "transform=rephoto",
      "target": "decision:gpt-6-luna",
      "value": 0.00019360000000000002,
      "lo": 0.00019360000000000002,
      "hi": 0.00019360000000000002,
      "n": 15,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "transform=rephoto",
      "target": "decision:gpt-6-luna",
      "value": 0.00019360000000000002,
      "lo": 0.00019360000000000002,
      "hi": 0.00019360000000000002,
      "n": 15,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "transform=rephoto",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "transform=rephoto",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "transform=rephoto",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "transform=resize",
      "target": "decision:gpt-6-luna",
      "value": 241,
      "lo": 241,
      "hi": 241,
      "n": 15,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "transform=resize",
      "target": "decision:gpt-6-luna",
      "value": 432.69999999999965,
      "lo": 432.69999999999965,
      "hi": 432.69999999999965,
      "n": 15,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "transform=resize",
      "target": "decision:gpt-6-luna",
      "value": 622.5399999999997,
      "lo": 622.5399999999997,
      "hi": 622.5399999999997,
      "n": 15,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "transform=resize",
      "target": "decision:gpt-6-luna",
      "value": 0.0000826,
      "lo": 0.0000826,
      "hi": 0.0000826,
      "n": 15,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "transform=resize",
      "target": "decision:gpt-6-luna",
      "value": 0.0000826,
      "lo": 0.0000826,
      "hi": 0.0000826,
      "n": 15,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "transform=resize",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "transform=resize",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "transform=resize",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "transform=rotate",
      "target": "decision:gpt-6-luna",
      "value": 377,
      "lo": 377,
      "hi": 377,
      "n": 15,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "transform=rotate",
      "target": "decision:gpt-6-luna",
      "value": 588.4999999999997,
      "lo": 588.4999999999997,
      "hi": 588.4999999999997,
      "n": 15,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "transform=rotate",
      "target": "decision:gpt-6-luna",
      "value": 781.6999999999998,
      "lo": 781.6999999999998,
      "hi": 781.6999999999998,
      "n": 15,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "transform=rotate",
      "target": "decision:gpt-6-luna",
      "value": 0.00019180000000000003,
      "lo": 0.00019180000000000003,
      "hi": 0.00019180000000000003,
      "n": 15,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "transform=rotate",
      "target": "decision:gpt-6-luna",
      "value": 0.00019180000000000003,
      "lo": 0.00019180000000000003,
      "hi": 0.00019180000000000003,
      "n": 15,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "transform=rotate",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "transform=rotate",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "transform=rotate",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p50_ms",
      "stratum": "transform=screenshot",
      "target": "decision:gpt-6-luna",
      "value": 342,
      "lo": 342,
      "hi": 342,
      "n": 15,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p95_ms",
      "stratum": "transform=screenshot",
      "target": "decision:gpt-6-luna",
      "value": 490.2,
      "lo": 490.2,
      "hi": 490.2,
      "n": 15,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "latency_p99_ms",
      "stratum": "transform=screenshot",
      "target": "decision:gpt-6-luna",
      "value": 509.24,
      "lo": 509.24,
      "hi": 509.24,
      "n": 15,
      "unit": "ms",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "cost_per_check_usd",
      "stratum": "transform=screenshot",
      "target": "decision:gpt-6-luna",
      "value": 0.000154,
      "lo": 0.000154,
      "hi": 0.000154,
      "n": 15,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "model_cost_per_check_usd",
      "stratum": "transform=screenshot",
      "target": "decision:gpt-6-luna",
      "value": 0.000154,
      "lo": 0.000154,
      "hi": 0.000154,
      "n": 15,
      "unit": "usd",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "error_rate",
      "stratum": "transform=screenshot",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "refusal_rate",
      "stratum": "transform=screenshot",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    },
    {
      "task": "operations",
      "metric": "timeout_rate",
      "stratum": "transform=screenshot",
      "target": "decision:gpt-6-luna",
      "value": 0,
      "lo": 0,
      "hi": 0.20388330103584862,
      "n": 15,
      "unit": "rate",
      "higherIsBetter": false
    }
  ],
  "findings": {
    "losses": [
      "extraction, documents with every labelled field right: Receipt Audit (Claude Haiku 5.5) 32.2% [24.6%–41%] (n=121) vs Claude Haiku 5.5, plain prompt 95% [89.6%–97.7%] (n=121).",
      "extraction, documents with every labelled field right: Receipt Audit on Gemini 3.8 Flash 32.5% [24.8%–41.3%] (n=121) vs Claude Haiku 5.5, plain prompt 95% [89.6%–97.7%] (n=121).",
      "extraction, total read right: Receipt Audit (Claude Haiku 5.5) 33.3% [25.6%–42.1%] (n=121) vs Claude Haiku 5.5, plain prompt 99.4% [95.9%–99.9%] (n=121).",
      "extraction, total read right: Receipt Audit on Gemini 3.8 Flash 33.3% [25.6%–42.1%] (n=121) vs Claude Haiku 5.5, plain prompt 99.4% [95.9%–99.9%] (n=121).",
      "extraction, line-item F1: Receipt Audit (Claude Haiku 5.5) 48.8% [46.4%–50%] (n=20) vs Claude Haiku 5.5, plain prompt 96.1% [90.5%–100%] (n=20).",
      "extraction, line-item F1: Receipt Audit on Gemini 3.8 Flash 43% [36.3%–48.3%] (n=20) vs Claude Haiku 5.5, plain prompt 96.1% [90.5%–100%] (n=20).",
      "decisions, same decision in every repetition: Receipt Audit (Claude Haiku 5.5) 79.3% [71.3%–85.6%] (n=121) vs OpenAI decision model (gpt-6-luna), asked if it's fake 100% [96.9%–100%] (n=121).",
      "decisions, same decision in every repetition: Receipt Audit on Gemini 3.8 Flash 75.2% [66.8%–82%] (n=121) vs Claude Haiku 5.5, plain prompt 90.9% [84.5%–94.8%] (n=121).",
      "decisions, same decision in every repetition: Receipt Audit on Gemini 3.8 Flash 75.2% [66.8%–82%] (n=121) vs OpenAI decision model (gpt-6-luna), asked if it's fake 100% [96.9%–100%] (n=121).",
      "operations, cost per check to the caller: Receipt Audit (Claude Haiku 5.5) $0.016 (n=121) vs Claude Haiku 5.5, plain prompt $0.00087 (n=453).",
      "operations, cost per check to the caller: Receipt Audit (Claude Haiku 5.5) $0.016 (n=121) vs Gemini 3.8 Flash, plain prompt $0.0044 (n=98).",
      "operations, cost per check to the caller: Receipt Audit (Claude Haiku 5.5) $0.016 (n=121) vs OpenAI decision model (gpt-6-luna), asked if it's fake $0.00017 (n=453).",
      "operations, cost per check to the caller: Receipt Audit on Gemini 3.8 Flash $0.016 (n=150) vs Claude Haiku 5.5, plain prompt $0.00087 (n=453).",
      "operations, cost per check to the caller: Receipt Audit on Gemini 3.8 Flash $0.016 (n=150) vs Gemini 3.8 Flash, plain prompt $0.0044 (n=98).",
      "operations, cost per check to the caller: Receipt Audit on Gemini 3.8 Flash $0.016 (n=150) vs OpenAI decision model (gpt-6-luna), asked if it's fake $0.00017 (n=453).",
      "operations, model cost per check: Receipt Audit (Claude Haiku 5.5) $0.0011 (n=121) vs Claude Haiku 5.5, plain prompt $0.00087 (n=453).",
      "operations, model cost per check: Receipt Audit (Claude Haiku 5.5) $0.0011 (n=121) vs OpenAI decision model (gpt-6-luna), asked if it's fake $0.00017 (n=453).",
      "operations, model cost per check: Receipt Audit on Gemini 3.8 Flash $0.0044 (n=150) vs Claude Haiku 5.5, plain prompt $0.00087 (n=453).",
      "operations, model cost per check: Receipt Audit on Gemini 3.8 Flash $0.0044 (n=150) vs OpenAI decision model (gpt-6-luna), asked if it's fake $0.00017 (n=453).",
      "operations, median latency: Receipt Audit (Claude Haiku 5.5) 4,791 ms (n=121) vs OpenAI decision model (gpt-6-luna), asked if it's fake 349 ms (n=453).",
      "operations, median latency: Receipt Audit on Gemini 3.8 Flash 4,890 ms (n=150) vs OpenAI decision model (gpt-6-luna), asked if it's fake 349 ms (n=453)."
    ],
    "wins": [
      "extraction, documents with every labelled field right: Receipt Audit (Claude Haiku 5.5) 32.2% [24.6%–41%] (n=121); OpenAI decision model (gpt-6-luna), asked if it's fake can't do this task.",
      "extraction, documents with every labelled field right: Receipt Audit on Gemini 3.8 Flash 32.5% [24.8%–41.3%] (n=121); OpenAI decision model (gpt-6-luna), asked if it's fake can't do this task.",
      "extraction, total read right: Receipt Audit (Claude Haiku 5.5) 33.3% [25.6%–42.1%] (n=121); OpenAI decision model (gpt-6-luna), asked if it's fake can't do this task.",
      "extraction, total read right: Receipt Audit on Gemini 3.8 Flash 33.3% [25.6%–42.1%] (n=121); OpenAI decision model (gpt-6-luna), asked if it's fake can't do this task.",
      "extraction, line-item F1: Receipt Audit (Claude Haiku 5.5) 48.8% [46.4%–50%] (n=20); OpenAI decision model (gpt-6-luna), asked if it's fake can't do this task.",
      "extraction, line-item F1: Receipt Audit on Gemini 3.8 Flash 43% [36.3%–48.3%] (n=20); OpenAI decision model (gpt-6-luna), asked if it's fake can't do this task.",
      "authenticity, ROC AUC, genuine vs fake: Receipt Audit (Claude Haiku 5.5) 0.870 [0.789–0.942] (n=121) vs OpenAI decision model (gpt-6-luna), asked if it's fake 0.648 [0.536–0.755] (n=121).",
      "authenticity, ROC AUC, genuine vs fake: Receipt Audit on Gemini 3.8 Flash 0.908 [0.837–0.966] (n=121) vs OpenAI decision model (gpt-6-luna), asked if it's fake 0.648 [0.536–0.755] (n=121).",
      "authenticity, fakes caught at a 5% false-flag rate: Receipt Audit (Claude Haiku 5.5) 77.5% [52.5%–90%] (n=40) vs OpenAI decision model (gpt-6-luna), asked if it's fake 30% [17.5%–47.5%] (n=40).",
      "authenticity, fakes caught at a 5% false-flag rate: Receipt Audit on Gemini 3.8 Flash 82.5% [67.5%–92.5%] (n=40) vs OpenAI decision model (gpt-6-luna), asked if it's fake 30% [17.5%–47.5%] (n=40).",
      "authenticity, fakes whose amounts add up, caught at a 5% false-flag rate (arithmetic=consistent): Receipt Audit (Claude Haiku 5.5) 72.7% [50%–90.9%] (n=22) vs Claude Haiku 5.5, plain prompt 22.7% [4.5%–45.5%] (n=22).",
      "authenticity, fakes whose amounts add up, caught at a 5% false-flag rate (arithmetic=consistent): Receipt Audit (Claude Haiku 5.5) 72.7% [50%–90.9%] (n=22) vs OpenAI decision model (gpt-6-luna), asked if it's fake 27.3% [9.1%–45.5%] (n=22).",
      "authenticity, fakes whose amounts add up, caught at a 5% false-flag rate (arithmetic=consistent): Receipt Audit on Gemini 3.8 Flash 77.3% [54.5%–95.5%] (n=22) vs Claude Haiku 5.5, plain prompt 22.7% [4.5%–45.5%] (n=22).",
      "authenticity, fakes whose amounts add up, caught at a 5% false-flag rate (arithmetic=consistent): Receipt Audit on Gemini 3.8 Flash 77.3% [54.5%–95.5%] (n=22) vs OpenAI decision model (gpt-6-luna), asked if it's fake 27.3% [9.1%–45.5%] (n=22).",
      "duplicates, resubmitted copies recognised: Receipt Audit (Claude Haiku 5.5) 0% [0%–11.4%] (n=30); Claude Haiku 5.5, plain prompt can't do this task.",
      "duplicates, resubmitted copies recognised: Receipt Audit (Claude Haiku 5.5) 0% [0%–11.4%] (n=30); Gemini 3.8 Flash, plain prompt can't do this task.",
      "duplicates, resubmitted copies recognised: Receipt Audit (Claude Haiku 5.5) 0% [0%–11.4%] (n=30); OpenAI decision model (gpt-6-luna), asked if it's fake can't do this task.",
      "duplicates, resubmitted copies recognised: Receipt Audit on Gemini 3.8 Flash 32.2% [18.4%–50.1%] (n=30); Claude Haiku 5.5, plain prompt can't do this task.",
      "duplicates, resubmitted copies recognised: Receipt Audit on Gemini 3.8 Flash 32.2% [18.4%–50.1%] (n=30); Gemini 3.8 Flash, plain prompt can't do this task.",
      "duplicates, resubmitted copies recognised: Receipt Audit on Gemini 3.8 Flash 32.2% [18.4%–50.1%] (n=30); OpenAI decision model (gpt-6-luna), asked if it's fake can't do this task.",
      "decisions, calibration error of the risk score: Receipt Audit on Gemini 3.8 Flash 7.6% [4.7%–12.1%] (n=121) vs OpenAI decision model (gpt-6-luna), asked if it's fake 18.3% [14.5%–21.8%] (n=121).",
      "proof, fakes caught with verifiable evidence: Receipt Audit (Claude Haiku 5.5) 16.7% [8.2%–31%] (n=40); Claude Haiku 5.5, plain prompt can't do this task.",
      "proof, fakes caught with verifiable evidence: Receipt Audit (Claude Haiku 5.5) 16.7% [8.2%–31%] (n=40); Gemini 3.8 Flash, plain prompt can't do this task.",
      "proof, fakes caught with verifiable evidence: Receipt Audit (Claude Haiku 5.5) 16.7% [8.2%–31%] (n=40); OpenAI decision model (gpt-6-luna), asked if it's fake can't do this task.",
      "proof, fakes caught with verifiable evidence: Receipt Audit on Gemini 3.8 Flash 20% [10.5%–34.8%] (n=40); Claude Haiku 5.5, plain prompt can't do this task.",
      "proof, fakes caught with verifiable evidence: Receipt Audit on Gemini 3.8 Flash 20% [10.5%–34.8%] (n=40); Gemini 3.8 Flash, plain prompt can't do this task.",
      "proof, fakes caught with verifiable evidence: Receipt Audit on Gemini 3.8 Flash 20% [10.5%–34.8%] (n=40); OpenAI decision model (gpt-6-luna), asked if it's fake can't do this task.",
      "operations, model cost per check: Receipt Audit (Claude Haiku 5.5) $0.0011 (n=121) vs Gemini 3.8 Flash, plain prompt $0.0044 (n=98).",
      "operations, median latency: Receipt Audit (Claude Haiku 5.5) 4,791 ms (n=121) vs Claude Haiku 5.5, plain prompt 8,424 ms (n=453).",
      "operations, median latency: Receipt Audit (Claude Haiku 5.5) 4,791 ms (n=121) vs Gemini 3.8 Flash, plain prompt 6,523 ms (n=98).",
      "operations, median latency: Receipt Audit on Gemini 3.8 Flash 4,890 ms (n=150) vs Claude Haiku 5.5, plain prompt 8,424 ms (n=453).",
      "operations, median latency: Receipt Audit on Gemini 3.8 Flash 4,890 ms (n=150) vs Gemini 3.8 Flash, plain prompt 6,523 ms (n=98)."
    ],
    "ties": [
      "extraction, documents with every labelled field right: Receipt Audit (Claude Haiku 5.5) 32.2% [24.6%–41%] (n=121) vs Gemini 3.8 Flash, plain prompt 25.9% [18.9%–34.4%] (n=121).",
      "extraction, documents with every labelled field right: Receipt Audit on Gemini 3.8 Flash 32.5% [24.8%–41.3%] (n=121) vs Gemini 3.8 Flash, plain prompt 25.9% [18.9%–34.4%] (n=121).",
      "extraction, total read right: Receipt Audit (Claude Haiku 5.5) 33.3% [25.6%–42.1%] (n=121) vs Gemini 3.8 Flash, plain prompt 27% [19.9%–35.5%] (n=121).",
      "extraction, total read right: Receipt Audit on Gemini 3.8 Flash 33.3% [25.6%–42.1%] (n=121) vs Gemini 3.8 Flash, plain prompt 27% [19.9%–35.5%] (n=121).",
      "extraction, line-item F1: Receipt Audit (Claude Haiku 5.5) 48.8% [46.4%–50%] (n=20) vs Gemini 3.8 Flash, plain prompt 40.4% [31.2%–47.5%] (n=20).",
      "extraction, line-item F1: Receipt Audit on Gemini 3.8 Flash 43% [36.3%–48.3%] (n=20) vs Gemini 3.8 Flash, plain prompt 40.4% [31.2%–47.5%] (n=20).",
      "authenticity, ROC AUC, genuine vs fake: Receipt Audit (Claude Haiku 5.5) 0.870 [0.789–0.942] (n=121) vs Claude Haiku 5.5, plain prompt 0.796 [0.696–0.886] (n=121).",
      "authenticity, ROC AUC, genuine vs fake: Receipt Audit (Claude Haiku 5.5) 0.870 [0.789–0.942] (n=121) vs Gemini 3.8 Flash, plain prompt 0.955 [0.898–1.000] (n=121).",
      "authenticity, ROC AUC, genuine vs fake: Receipt Audit on Gemini 3.8 Flash 0.908 [0.837–0.966] (n=121) vs Claude Haiku 5.5, plain prompt 0.796 [0.696–0.886] (n=121).",
      "authenticity, ROC AUC, genuine vs fake: Receipt Audit on Gemini 3.8 Flash 0.908 [0.837–0.966] (n=121) vs Gemini 3.8 Flash, plain prompt 0.955 [0.898–1.000] (n=121).",
      "authenticity, fakes caught at a 5% false-flag rate: Receipt Audit (Claude Haiku 5.5) 77.5% [52.5%–90%] (n=40) vs Claude Haiku 5.5, plain prompt 57.5% [40%–72.5%] (n=40).",
      "authenticity, fakes caught at a 5% false-flag rate: Receipt Audit (Claude Haiku 5.5) 77.5% [52.5%–90%] (n=40) vs Gemini 3.8 Flash, plain prompt 95% [87.5%–100%] (n=40).",
      "authenticity, fakes caught at a 5% false-flag rate: Receipt Audit on Gemini 3.8 Flash 82.5% [67.5%–92.5%] (n=40) vs Claude Haiku 5.5, plain prompt 57.5% [40%–72.5%] (n=40).",
      "authenticity, fakes caught at a 5% false-flag rate: Receipt Audit on Gemini 3.8 Flash 82.5% [67.5%–92.5%] (n=40) vs Gemini 3.8 Flash, plain prompt 95% [87.5%–100%] (n=40).",
      "authenticity, fakes whose amounts add up, caught at a 5% false-flag rate (arithmetic=consistent): Receipt Audit (Claude Haiku 5.5) 72.7% [50%–90.9%] (n=22) vs Gemini 3.8 Flash, plain prompt 90.9% [77.3%–100%] (n=22).",
      "authenticity, fakes whose amounts add up, caught at a 5% false-flag rate (arithmetic=consistent): Receipt Audit on Gemini 3.8 Flash 77.3% [54.5%–95.5%] (n=22) vs Gemini 3.8 Flash, plain prompt 90.9% [77.3%–100%] (n=22).",
      "authenticity, genuine receipts flagged at risk ≥ 0.5: Receipt Audit (Claude Haiku 5.5) 0.8% [0.1%–6%] (n=81) vs Claude Haiku 5.5, plain prompt 1.6% [0.4%–7.3%] (n=81).",
      "authenticity, genuine receipts flagged at risk ≥ 0.5: Receipt Audit (Claude Haiku 5.5) 0.8% [0.1%–6%] (n=81) vs Gemini 3.8 Flash, plain prompt 0.4% [0%–5.3%] (n=81).",
      "authenticity, genuine receipts flagged at risk ≥ 0.5: Receipt Audit (Claude Haiku 5.5) 0.8% [0.1%–6%] (n=81) vs OpenAI decision model (gpt-6-luna), asked if it's fake 0% [0%–4.5%] (n=81).",
      "authenticity, genuine receipts flagged at risk ≥ 0.5: Receipt Audit on Gemini 3.8 Flash 0.8% [0.1%–6%] (n=81) vs Claude Haiku 5.5, plain prompt 1.6% [0.4%–7.3%] (n=81).",
      "authenticity, genuine receipts flagged at risk ≥ 0.5: Receipt Audit on Gemini 3.8 Flash 0.8% [0.1%–6%] (n=81) vs Gemini 3.8 Flash, plain prompt 0.4% [0%–5.3%] (n=81).",
      "authenticity, genuine receipts flagged at risk ≥ 0.5: Receipt Audit on Gemini 3.8 Flash 0.8% [0.1%–6%] (n=81) vs OpenAI decision model (gpt-6-luna), asked if it's fake 0% [0%–4.5%] (n=81).",
      "decisions, same decision in every repetition: Receipt Audit (Claude Haiku 5.5) 79.3% [71.3%–85.6%] (n=121) vs Claude Haiku 5.5, plain prompt 90.9% [84.5%–94.8%] (n=121).",
      "decisions, same decision in every repetition: Receipt Audit (Claude Haiku 5.5) 79.3% [71.3%–85.6%] (n=121) vs Gemini 3.8 Flash, plain prompt 86.8% [79.6%–91.7%] (n=121).",
      "decisions, same decision in every repetition: Receipt Audit on Gemini 3.8 Flash 75.2% [66.8%–82%] (n=121) vs Gemini 3.8 Flash, plain prompt 86.8% [79.6%–91.7%] (n=121).",
      "decisions, calibration error of the risk score: Receipt Audit (Claude Haiku 5.5) 9.2% [6.8%–15%] (n=121) vs Claude Haiku 5.5, plain prompt 10.9% [8.6%–15.2%] (n=121).",
      "decisions, calibration error of the risk score: Receipt Audit (Claude Haiku 5.5) 9.2% [6.8%–15%] (n=121) vs Gemini 3.8 Flash, plain prompt 3.5% [1.7%–7.1%] (n=121).",
      "decisions, calibration error of the risk score: Receipt Audit (Claude Haiku 5.5) 9.2% [6.8%–15%] (n=121) vs OpenAI decision model (gpt-6-luna), asked if it's fake 18.3% [14.5%–21.8%] (n=121).",
      "decisions, calibration error of the risk score: Receipt Audit on Gemini 3.8 Flash 7.6% [4.7%–12.1%] (n=121) vs Claude Haiku 5.5, plain prompt 10.9% [8.6%–15.2%] (n=121).",
      "decisions, calibration error of the risk score: Receipt Audit on Gemini 3.8 Flash 7.6% [4.7%–12.1%] (n=121) vs Gemini 3.8 Flash, plain prompt 3.5% [1.7%–7.1%] (n=121).",
      "decisions, expected cost of decisions per 1,000 receipts at risk ≥ 0.5: Receipt Audit (Claude Haiku 5.5) $946.852 [$775.000–$1171.296] (n=121) vs Claude Haiku 5.5, plain prompt $802.037 [$494.259–$1195.806] (n=121).",
      "decisions, expected cost of decisions per 1,000 receipts at risk ≥ 0.5: Receipt Audit (Claude Haiku 5.5) $946.852 [$775.000–$1171.296] (n=121) vs Gemini 3.8 Flash, plain prompt $944.259 [$841.667–$1107.778] (n=121).",
      "decisions, expected cost of decisions per 1,000 receipts at risk ≥ 0.5: Receipt Audit (Claude Haiku 5.5) $946.852 [$775.000–$1171.296] (n=121) vs OpenAI decision model (gpt-6-luna), asked if it's fake $775.000 [$625.000–$900.000] (n=121).",
      "decisions, expected cost of decisions per 1,000 receipts at risk ≥ 0.5: Receipt Audit on Gemini 3.8 Flash $905.185 [$741.667–$1121.296] (n=121) vs Claude Haiku 5.5, plain prompt $802.037 [$494.259–$1195.806] (n=121).",
      "decisions, expected cost of decisions per 1,000 receipts at risk ≥ 0.5: Receipt Audit on Gemini 3.8 Flash $905.185 [$741.667–$1121.296] (n=121) vs Gemini 3.8 Flash, plain prompt $944.259 [$841.667–$1107.778] (n=121).",
      "decisions, expected cost of decisions per 1,000 receipts at risk ≥ 0.5: Receipt Audit on Gemini 3.8 Flash $905.185 [$741.667–$1121.296] (n=121) vs OpenAI decision model (gpt-6-luna), asked if it's fake $775.000 [$625.000–$900.000] (n=121).",
      "evidence, edited fakes flagged with the edited field named: Receipt Audit (Claude Haiku 5.5) 11.7% [3.6%–32.1%] (n=20) vs Claude Haiku 5.5, plain prompt 41.7% [23.2%–62.8%] (n=20).",
      "evidence, edited fakes flagged with the edited field named: Receipt Audit (Claude Haiku 5.5) 11.7% [3.6%–32.1%] (n=20) vs Gemini 3.8 Flash, plain prompt 0% [0%–16.1%] (n=20).",
      "evidence, edited fakes flagged with the edited field named: Receipt Audit (Claude Haiku 5.5) 11.7% [3.6%–32.1%] (n=20) vs OpenAI decision model (gpt-6-luna), asked if it's fake 0% [0%–16.1%] (n=20).",
      "evidence, edited fakes flagged with the edited field named: Receipt Audit on Gemini 3.8 Flash 15% [5.2%–36%] (n=20) vs Claude Haiku 5.5, plain prompt 41.7% [23.2%–62.8%] (n=20).",
      "evidence, edited fakes flagged with the edited field named: Receipt Audit on Gemini 3.8 Flash 15% [5.2%–36%] (n=20) vs Gemini 3.8 Flash, plain prompt 0% [0%–16.1%] (n=20).",
      "evidence, edited fakes flagged with the edited field named: Receipt Audit on Gemini 3.8 Flash 15% [5.2%–36%] (n=20) vs OpenAI decision model (gpt-6-luna), asked if it's fake 0% [0%–16.1%] (n=20).",
      "operations, model cost per check: Receipt Audit on Gemini 3.8 Flash $0.0044 (n=150) vs Gemini 3.8 Flash, plain prompt $0.0044 (n=98)."
    ]
  },
  "datasets": [
    {
      "name": "SROIE (ICDAR 2019 scanned receipts), Voxel51/scanned_receipts mirror",
      "license": "CC-BY-4.0",
      "url": "https://huggingface.co/datasets/Voxel51/scanned_receipts",
      "cases": 98
    },
    {
      "name": "RA-Bench fakes from SROIE",
      "license": "CC-BY-4.0",
      "url": "https://huggingface.co/datasets/Voxel51/scanned_receipts",
      "cases": 20
    },
    {
      "name": "RA-Bench generated fakes",
      "license": "CC-BY-4.0",
      "url": null,
      "cases": 20
    },
    {
      "name": "Wikimedia Commons (Naha Mama Pavilionz)",
      "license": "CC0",
      "url": "https://commons.wikimedia.org/wiki/File:JP_%E6%97%A5%E6%9C%AC_Japan_%E6%B2%96%E7%B9%A9_OKINAWA_Family_Cuisine_Restaurant_January_2025_R12S_01.jpg",
      "cases": 2
    },
    {
      "name": "Wikimedia Commons (NORTLAUKAU 26088 JP)",
      "license": "CC0",
      "url": "https://commons.wikimedia.org/wiki/File:JP_%E6%97%A5%E6%9C%AC_Japan_%E6%B2%96%E7%B9%A9_OKINAWA_Route_253_%E6%B5%A6%E8%A5%BF_Chome_Yoshinoya_Restaurant_%E7%89%9B%E4%B8%BC%E9%A3%AF_Beef_slice_rice_bowl_February_2026_N13P_05.jpg",
      "cases": 2
    },
    {
      "name": "Wikimedia Commons (Solomon203)",
      "license": "CC BY-SA 4.0",
      "url": "https://commons.wikimedia.org/wiki/File:FamilyMart_Keelung_Baifu_Store_e-invoice_WJ90244062.jpg",
      "cases": 2
    },
    {
      "name": "Wikimedia Commons (Aw58)",
      "license": "CC BY-SA 4.0",
      "url": "https://commons.wikimedia.org/wiki/File:Paragon_fiskalny_-_12.07.2024.jpg",
      "cases": 1
    },
    {
      "name": "Wikimedia Commons (Grandmaster Huon)",
      "license": "Public domain",
      "url": "https://commons.wikimedia.org/wiki/File:Grocery_Store_Receipt_in_Vienna.jpg",
      "cases": 1
    },
    {
      "name": "Wikimedia Commons (Lantus)",
      "license": "Public domain",
      "url": "https://commons.wikimedia.org/wiki/File:Migros_Kassenzettel_2025.pdf",
      "cases": 1
    },
    {
      "name": "Wikimedia Commons (LIman TADU Foo RUBB)",
      "license": "CC0",
      "url": "https://commons.wikimedia.org/wiki/File:SZ_%E6%B7%B1%E5%9C%B3_Shenzhen_%E9%BE%8D%E5%B4%97_Longgang_%E4%BA%94%E5%92%8C%E8%B7%AF_118_Wuhe_Road_%E6%B5%B7%E8%BE%BE%E5%92%8C%E6%88%90%E8%B4%AD%E7%89%A9%E4%B8%AD%E5%BF%83_HaiDa_HeCheng_Shopping_Mall_shop_%E6%B5%81%E6%B5%AA%E6%B3%A1%E6%B3%A1%E7%83%A4%E8%82%89_LiuLang_PaoPao_Korean_BBQ_Buffet_Restaurant_bill_n_rules_November_2025_R12S_01.jpg",
      "cases": 1
    },
    {
      "name": "Wikimedia Commons (Sakurakat)",
      "license": "CC0",
      "url": "https://commons.wikimedia.org/wiki/File:Bill_Vishwanand_2026-01-28.jpg",
      "cases": 1
    },
    {
      "name": "Wikimedia Commons (TaiZiThee 832)",
      "license": "CC0",
      "url": "https://commons.wikimedia.org/wiki/File:SZ_%E6%B7%B1%E5%9C%B3_Shenzhen_%E5%AF%B6%E5%AE%89%E4%B8%AD%E5%BF%83_BaoAn_%E5%89%B5%E6%A5%AD%E4%B8%80%E8%B7%AF_Chuangye_1st_Road_mall_AEON_Supermarket_bill_December_2024_R12S_01.jpg",
      "cases": 1
    },
    {
      "name": "Wikimedia Commons (YEING Luwem Hai Wangs)",
      "license": "CC0",
      "url": "https://commons.wikimedia.org/wiki/File:SZ_%E6%B7%B1%E5%9C%B3_Shenzhen_%E7%A6%8F%E7%94%B0_Futian_%E9%87%91%E7%94%B0%E8%B7%AF_Jintian_Road_%E4%B8%83%E9%AE%AE%E8%B6%85%E5%B8%82_Seven_Fresh_Supermarket_checkout_bill_February_2024_R12S_20.jpg",
      "cases": 1
    }
  ],
  "limitations": [
    "Preliminary: small samples, so intervals are wide. Read the intervals, not the point estimates.",
    "Public cases come from datasets that are probably in the models' training data.",
    "The DIY baselines use one plain prompt; a stronger prompt would do better. A strong-prompt baseline comes later.",
    "DIY targets have no memory across checks and no deterministic checks: those tasks are marked not supported, not scored as zero.",
    "Our engine runs in process, so its latency has no network hop to receiptaudit.com; DIY latency includes the round trip to the model.",
    "Costs are model tokens at list price on the price date; no infrastructure cost is included.",
    "When the Claude baseline's config says transport claude-code-cli, it ran through Claude Code's CLI on a subscription: thinking is adaptive and can't be fully turned off, its cost is what the same tokens cost through the API, and its latency includes the CLI's start-up.",
    "The duplicate network starts empty for each repetition: it only knows the run's own earlier cases, not production traffic."
  ],
  "examples": [],
  "methodology": "RA-Bench 0.2.0. These definitions, thresholds and exclusions were written before the run and change only in a new suite version.\n\nWho and what. Receipt Audit runs this benchmark on its own product. Every target gets the same cases, the same files, the same timeout (180 s per attempt by default) and the same scoring; only the target changes. A target is our engine (\"self\", in process, with the duplicate network on an empty in-memory backend that lives for one repetition) or an agent's own model asked with one plain prompt (\"diy\"). DIY targets get the same file preparation as our engine (upright, at most 3072 px and 3.75 MB) so that API size limits don't count against them. Transport failures (network errors, HTTP 408, 429 and 5xx) are retried up to 3 times with exponential backoff; nothing else is retried, and a wrong answer is a result. Models are pinned: no fallback to another model.\n\nUnits and intervals. A case run several times counts once, with the share of its runs that succeeded. Rates carry a 95% Wilson interval over cases (operations rates: over calls). ROC AUC, detection at a fixed false-flag rate and line-item F1 carry a 95% percentile bootstrap interval (2000 resamples, genuine receipts and fakes resampled separately, seeded per row so a re-scored run gives the same interval). Latency and cost have no interval. Every row reports n. Rows are given for all cases and for every stratum (a \"dimension=value\" tag on the case, or a label-derived group) with at least 10 cases.\n\nWhich cases count. Extraction, checks and authenticity use base cases only: cases that aren't a transform of another case (no lineage). Duplicate transforms count only for the duplicates task. Operations use every call.\n\nExtraction. Scored only on fields the label covers; a field labelled null means the document doesn't show it, and only a null answer is right. Merchant name: lowercase, accents stripped, \"&\" read as \"and\", punctuation dropped, then a match at edit-distance similarity ≥ 0.9. Date: the exact ISO date (YYYY-MM-DD). Subtotal, tax total and total: within 0.01. Currency: the exact ISO 4217 code. Receipt number: equal after keeping only letters and digits, lowercased. \"All fields right\": every labelled scalar field of the document right in that run (line items excluded). Line-item F1: labelled and extracted items paired one to one when descriptions are at least 0.8 alike (same normalisation) and totals match within 0.01, micro-averaged over all documents. Valid output: the call succeeded and returned readable fields. A failed call or an unreadable answer gets every field wrong.\n\nChecks. totals_add_up is scored against the label \"the printed amounts add up\". A positive is a document whose amounts don't add up; the check flags it when it reports failed. Recall over documents that don't add up; precision over documents it flagged; false alarms over genuine documents that do add up; coverage is the share of documents where the check ran (a check that couldn't run isn't output). Targets without deterministic checks report this task as not supported.\n\nAuthenticity (genuine vs fake). Higher risk means more likely fake. ROC AUC uses each target's risk as the score; a result without a risk scores 0.5 (no opinion), ties count half. Detection at a 1%, 5% and 10% false-flag rate: the share of fakes flagged by the lowest threshold that flags at most that share of genuine receipts (ties at the threshold are all flagged). At the documented thresholds risk ≥ 0.5 and risk ≥ 0.7: the share of fakes flagged (detection) and of genuine receipts flagged (false flags); a result without a risk is never flagged. Strata from the labels: each fake class (\"class=ai_full\", …) and fakes whose amounts add up or don't (\"arithmetic=consistent\", \"arithmetic=inconsistent\"), each compared with every genuine receipt; and provenance metadata (\"provenance=intact\", \"stripped\", \"none\"), genuine receipts and fakes alike.\n\nDuplicates. Every case is submitted by a different account, so a repeat is \"someone else submitted this receipt before\". Recall: the share of duplicate transforms (re-encoded, cropped, re-shot copies of an earlier case) the target recognised as seen before; exact recall counts only matches on the identical image hash. False matches: the share of distinct receipts reported as seen before, for hard negatives (different receipts that look alike: same merchant, template or day; \"false_match_rate\") and for the other genuine receipts (\"false_match_rate_genuine\"). Originals always run before their transforms. Targets without memory across checks (DIY prompts) report the task as not supported, not as zero.\n\nDecisions. A decision is the yes/no a caller acts on: flagged when risk ≥ 0.5. Consistency: over base cases run at least twice, the share whose decision is the same in every repetition; total consistency, the share whose extracted total is the same (within 0.01, or absent) in every repetition. Calibration error: expected calibration error of risk against \"is fake\" over every run with a risk, 10 equal-width bins, weighted by run count (lower is better; it depends on this set's share of fakes). Expected cost per 1,000 receipts at risk ≥ 0.5 and ≥ 0.7: 1000 × ((1 − p) × false-flag rate × C_review + p × (1 − detection) × C_miss). Defaults, each from a cited, graded source (research/41-benchmark-cost-model.md): p = 1% of submitted receipts are fake (SAP Concur's share of receipts flagged as AI-generated; no study measures the true share, so this is the best available assumption, not a measurement); C_miss = $100, the face value of a missed fake (AppZen's mean AI-generated fake, 12 months to May 2026); C_review = $17, one false flag's review (18 minutes, GBTA, at a loaded auditor wage of $57.47/h, BLS). At these defaults one point of false-flag rate costs as much as about 17 points of detection. The sweep varies one assumption at a time over its published spread (\"assumption=…\": fakes 0.1% and 5%; a missed fake $32 and $182; a review $8 and $73). Approving everything costs 1000 × p × C_miss ($1000 at the defaults), reported as its own row: a detector above it loses money under those assumptions. The formula counts only mistakes (it treats reviewing a fake that was rightly flagged as free) and leaves out unmeasured costs: an employee's time and trust after a false flag, and repeat fraud or tax penalties after a miss. This set has far more fakes than real submissions, so the cost re-weights the measured rates by p rather than using the set's own mix.\n\nEvidence. A yes/no is easy to get from any model; a reviewer also needs to know what is wrong. Edited field named: over fakes whose label names the edited field (AI-inpainted and hand-edited fakes), the share of runs that flag the receipt and whose first 3 reasons point at that field. A reason points at a field through an explicit \"@field\" tag, a deterministic check about it (an arithmetic check points at the total and subtotal), or, for free-text reasons, a pre-registered keyword (total, amount due; subtotal; tax, VAT, GST; date; merchant, store, vendor, logo, name; item, line, quantity, price).\n\nVerifiable evidence. A finding is proof of tampering when a caller can check it without trusting a model and it shows the document was generated or changed: an AI-generator, AI source-type or receipt/invoice-generator tag read from the file, PDF modification records, a thumbnail that doesn't match the image, a photo taken before the printed purchase date, printed amounts that don't add up (with the numbers), or a duplicate-network match. Other file facts (an editor or office tool, missing camera data) and plausibility or policy checks (tax rate, the receipt's age, currency) are real but don't show tampering, and don't count. A model's visual judgement is an opinion, however confident. Fakes caught with proof: the share of fakes flagged with at least one proof among their reasons. Genuine receipts with proof: the guard against false proof (lower is better). Only targets that can return verifiable findings report this task; a plain prompt returns opinions by construction and reports it as not supported.\n\nOperations. Latency: wall time of the final attempt of each successful call, measured by the harness (p50, p95, p99). Cost per check is what the caller pays for one check: for Receipt Audit, the list price of its smallest credit pack per check; for a DIY baseline, its model tokens × the AI Gateway list price on the run's price date. Model cost per check is the model tokens × that list price for every target, our engine's own model call included (infrastructure not included). Both are averaged over successful calls. Error, refusal and timeout rates over all calls.\n\nFindings. Each of our targets is compared with each baseline on the headline rows of every task (stratum \"all\"): a win or a loss only when the 95% intervals don't overlap, otherwise a tie; for latency and cost, which have no interval, only when one value is at least 1.2 times the other. A comparison resting on fewer than 10 cases (or calls) on either side is always a tie. A task a baseline can't do by construction is reported as such, not as a score of zero.",
  "reproduce": {
    "repo": null,
    "command": "cd apps/receipt-check && nub run bench run --cases benchmarks/receipt-audit/heldout-cases.jsonl --split private --targets self,self-gemini,diy:claude-haiku-5.5,diy:gemini-3.8-flash,decision:gpt-6-luna --reps 3"
  },
  "history": [
    {
      "runId": "20261012T230424Z-fc67c6",
      "ranAt": "2026-10-12T23:18:58.299Z",
      "url": "https://receiptaudit.com/benchmarks/runs/20261012T230424Z-fc67c6.json"
    },
    {
      "runId": "20261012T223232Z-63be0d",
      "ranAt": "2026-10-12T22:49:04.881Z",
      "url": "https://receiptaudit.com/benchmarks/runs/20261012T223232Z-63be0d.json"
    },
    {
      "runId": "20261012T203648Z-44a89a",
      "ranAt": "2026-10-12T20:47:27.588Z",
      "url": "https://receiptaudit.com/benchmarks/runs/20261012T203648Z-44a89a.json"
    }
  ],
  "incomplete": {
    "reason": "44% of calls failed with an error",
    "failedCalls": 990,
    "totalCalls": 2265
  }
}