{
  "evidence": {
    "runId": "ac7e00c0-0e40-4927-9926-5af8c01171fe",
    "evaluatedAt": "2026-09-27T07:56:26.559Z",
    "packageVersion": "0.7.0",
    "recipeFingerprint": "f6f9cdae49cfd815b8389061f1c8bc6ba27a6a62517fbdcb184d1487f12f8047",
    "datasetFingerprint": "e28ffa155fc17f54bbfafc1a3c65516fa236634196b01dd268d701104011bf1f",
    "split": "held-out",
    "mode": "live",
    "policy": {
      "minConfidence": 0.8,
      "selectedOn": {
        "runId": "0d697590-4e91-4963-8d35-d45b3456c202",
        "datasetFingerprint": "ea60e52ab2ae50bc0a745091a78813fc4592d5b1e5712b400f40895e9e51384f",
        "recipeFingerprint": "f6f9cdae49cfd815b8389061f1c8bc6ba27a6a62517fbdcb184d1487f12f8047",
        "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": 25
      }
    ]
  },
  "recipe": "buying-intent",
  "model": "jev-1.13.0",
  "models": [
    "jev-1.13.0"
  ],
  "cases": 25,
  "completed": 25,
  "failed": 0,
  "correct": 24,
  "wrong": 1,
  "ready": 24,
  "review": 1,
  "accuracy": 0.96,
  "completedAccuracy": 0.96,
  "accuracyInterval95": [
    0.8046,
    0.9929
  ],
  "readyAccuracy": 0.958,
  "readyAccuracyInterval95": [
    0.7976,
    0.9926
  ],
  "reviewRate": 0.04,
  "contestedAccuracy": null,
  "adversarialAccuracy": null,
  "calibration": [
    {
      "confidence": "0-0.5",
      "cases": 0,
      "accuracy": null
    },
    {
      "confidence": "0.5-0.6",
      "cases": 1,
      "accuracy": 1
    },
    {
      "confidence": "0.6-0.7",
      "cases": 0,
      "accuracy": null
    },
    {
      "confidence": "0.7-0.8",
      "cases": 0,
      "accuracy": null
    },
    {
      "confidence": "0.8-0.9",
      "cases": 1,
      "accuracy": 1
    },
    {
      "confidence": "0.9-1",
      "cases": 23,
      "accuracy": 0.957
    }
  ],
  "thresholdEvaluation": "frozen-policy",
  "thresholds": [],
  "suggestedMinConfidence": null,
  "latencyMs": {
    "p50": 110.34,
    "p95": 163.93,
    "totalCaseTime": 2886.97
  },
  "usage": {
    "requests": 25,
    "input_tokens": 12080,
    "output_tokens": 425,
    "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": 0.00050736
  },
  "confusion": {
    "{\"intent\":\"committed\"}": {
      "total": 5,
      "correct": 5,
      "misses": {}
    },
    "{\"intent\":\"curious\"}": {
      "total": 5,
      "correct": 4,
      "misses": {
        "{\"intent\":\"none\"}": 1
      }
    },
    "{\"intent\":\"evaluating\"}": {
      "total": 5,
      "correct": 5,
      "misses": {}
    },
    "{\"intent\":\"none\"}": {
      "total": 5,
      "correct": 5,
      "misses": {}
    },
    "{\"intent\":\"ready\"}": {
      "total": 5,
      "correct": 5,
      "misses": {}
    }
  },
  "failures": [
    {
      "id": "featured-17-1",
      "expected": {
        "intent": "curious"
      },
      "actual": {
        "intent": "none"
      },
      "confidence": 0.95
    }
  ],
  "acceptance": {
    "minConfidence": 0.8,
    "minimumHeldOutReadyCases": 20,
    "minimumReadyAccuracy": 0.95,
    "maximumProviderFailures": 0,
    "met": true,
    "label": "Measured on these synthetic cases"
  }
}