{
  "evidence": {
    "runId": "dd3165ae-8e6c-4fd0-b372-a877188b903c",
    "evaluatedAt": "2026-09-27T07:56:33.689Z",
    "packageVersion": "0.7.0",
    "recipeFingerprint": "937fa33427abf5c862ed5c9f34919d1a5585a05b6cd7a722c857f0405f2dfb1d",
    "datasetFingerprint": "f5796d3fee2981671c27058fdc14bdf4761f519f03813154b50b9f72073b9843",
    "split": "held-out",
    "mode": "live",
    "policy": {
      "minConfidence": 0.8,
      "selectedOn": {
        "runId": "d0a2533c-0273-4f7d-aeb9-807ee79af896",
        "datasetFingerprint": "bbf1f8e57ebea24150cb4cc2c7bb371fc606adc093630fae41789b2411af4241",
        "recipeFingerprint": "937fa33427abf5c862ed5c9f34919d1a5585a05b6cd7a722c857f0405f2dfb1d",
        "model": "jev-1.13.0",
        "split": "development"
      }
    },
    "provenance": [
      {
        "method": "author-synthetic",
        "source": "AI-authored synthetic boundary cases for jev-recipes; labels have not had independent human review.",
        "cases": 40
      }
    ]
  },
  "recipe": "entity-match",
  "model": "jev-1.13.0",
  "models": [
    "jev-1.13.0"
  ],
  "cases": 40,
  "completed": 40,
  "failed": 0,
  "correct": 35,
  "wrong": 5,
  "ready": 30,
  "review": 10,
  "accuracy": 0.875,
  "completedAccuracy": 0.875,
  "accuracyInterval95": [
    0.7389,
    0.9454
  ],
  "readyAccuracy": 1,
  "readyAccuracyInterval95": [
    0.8865,
    1
  ],
  "reviewRate": 0.25,
  "contestedAccuracy": null,
  "adversarialAccuracy": null,
  "calibration": [
    {
      "confidence": "0-0.5",
      "cases": 0,
      "accuracy": null
    },
    {
      "confidence": "0.5-0.6",
      "cases": 7,
      "accuracy": 0.571
    },
    {
      "confidence": "0.6-0.7",
      "cases": 3,
      "accuracy": 0.333
    },
    {
      "confidence": "0.7-0.8",
      "cases": 0,
      "accuracy": null
    },
    {
      "confidence": "0.8-0.9",
      "cases": 0,
      "accuracy": null
    },
    {
      "confidence": "0.9-1",
      "cases": 30,
      "accuracy": 1
    }
  ],
  "thresholdEvaluation": "frozen-policy",
  "thresholds": [],
  "suggestedMinConfidence": null,
  "latencyMs": {
    "p50": 113.93,
    "p95": 166.59,
    "totalCaseTime": 4850.12
  },
  "usage": {
    "requests": 40,
    "input_tokens": 20859,
    "output_tokens": 800,
    "unknownResponses": 0,
    "failedRequests": 0,
    "models": [
      "jev-1.13.0"
    ]
  },
  "cost": {
    "model": "jev-1.13.0",
    "inputPerMillion": 0.042,
    "outputPerMillion": 0,
    "currency": "USD",
    "date": "2026-09-27",
    "source": "https://docs.typesafe.ai/models",
    "estimated": 0.00087608
  },
  "confusion": {
    "{\"verdict\":\"different\"}": {
      "total": 10,
      "correct": 5,
      "misses": {
        "{\"verdict\":\"same\"}": 5
      }
    },
    "{\"verdict\":\"same\"}": {
      "total": 30,
      "correct": 30,
      "misses": {}
    }
  },
  "failures": [
    {
      "id": "featured-17-different-id",
      "expected": {
        "verdict": "different"
      },
      "actual": {
        "verdict": "same"
      },
      "confidence": 0.62
    },
    {
      "id": "featured-19-different-id",
      "expected": {
        "verdict": "different"
      },
      "actual": {
        "verdict": "same"
      },
      "confidence": 0.55
    },
    {
      "id": "featured-22-different-id",
      "expected": {
        "verdict": "different"
      },
      "actual": {
        "verdict": "same"
      },
      "confidence": 0.6
    },
    {
      "id": "featured-23-different-id",
      "expected": {
        "verdict": "different"
      },
      "actual": {
        "verdict": "same"
      },
      "confidence": 0.52
    },
    {
      "id": "featured-24-different-id",
      "expected": {
        "verdict": "different"
      },
      "actual": {
        "verdict": "same"
      },
      "confidence": 0.58
    }
  ],
  "acceptance": {
    "minConfidence": 0.8,
    "minimumHeldOutReadyCases": 20,
    "minimumReadyAccuracy": 0.95,
    "maximumProviderFailures": 0,
    "met": true,
    "label": "Measured on these synthetic cases"
  }
}