{
  "measured": "2026-10-02",
  "sourceReport": "REPORT_QDECaR_Versus_Verdict_Retry_2026-10-01",
  "sourceSha256": "781d0bffc295a54a7009c83cef9c3d5736c2e3531ee189872aaacb08f8d7cc2d",
  "model": "Ternary Bonsai 2 27B PTQ1_0",
  "modelBytes": 5946648928,
  "modelSha256": "53107f530aa52eb00912263ab1ee29bd199261c87cd7b4ad4ca1318c1fe33ee3",
  "hardware": "M5 Max 128 GB inference; M4 Max grading",
  "formalEligible": false,
  "uniqueTasks": 237,
  "calls": 1120,
  "cohorts": {
    "hard148": {
      "n": 148,
      "arms": {
        "baseline": {
          "passed": 51,
          "n": 148,
          "input": 60747,
          "output": 78217,
          "tokens": 138964,
          "inferenceSeconds": 3843.7690247629507,
          "checkSeconds": 563.0049249140429,
          "baseline": "baseline"
        },
        "smart_sampled_baseline": {
          "passed": 42,
          "n": 148,
          "input": 60747,
          "output": 64499,
          "tokens": 125246,
          "inferenceSeconds": 3184.234167996008,
          "checkSeconds": 449.3073567949905,
          "baseline": "smart_sampled_baseline"
        },
        "smart_retry_T08": {
          "passed": 57,
          "n": 148,
          "input": 100743,
          "output": 108559,
          "tokens": 209302,
          "inferenceSeconds": 5274.254574039893,
          "checkSeconds": 991.719140791989,
          "baseline": "smart_sampled_baseline"
        },
        "oracle_one_resample": {
          "passed": 57,
          "n": 148,
          "input": 99847,
          "output": 123526,
          "tokens": 223373,
          "inferenceSeconds": 6108.637213808965,
          "checkSeconds": 1016.4475518769468,
          "baseline": "baseline"
        },
        "oracle_one_repair": {
          "passed": 65,
          "n": 148,
          "input": 167353,
          "output": 132180,
          "tokens": 299533,
          "inferenceSeconds": 6615.257056558898,
          "checkSeconds": 761.3125798279798,
          "baseline": "baseline"
        },
        "generated_one_repair": {
          "passed": 51,
          "n": 148,
          "input": 139546,
          "output": 149798,
          "tokens": 289344,
          "inferenceSeconds": 7487.889922763963,
          "checkSeconds": 902.21724554307,
          "baseline": "baseline"
        },
        "production_fast": {
          "passed": 51,
          "n": 148,
          "input": 135668,
          "output": 146668,
          "tokens": 282336,
          "inferenceSeconds": 7336.326229512968,
          "checkSeconds": 917.4863804580964,
          "baseline": "baseline"
        }
      }
    },
    "complete100": {
      "n": 100,
      "arms": {
        "baseline": {
          "passed": 57,
          "n": 100,
          "input": 38282,
          "output": 42107,
          "tokens": 80389,
          "inferenceSeconds": 1331.4562424578762,
          "checkSeconds": 124.85901183099486,
          "baseline": "baseline"
        },
        "smart_sampled_baseline": {
          "passed": 57,
          "n": 100,
          "input": 38282,
          "output": 33709,
          "tokens": 71991,
          "inferenceSeconds": 1092.8364078320446,
          "checkSeconds": 105.31598887502332,
          "baseline": "smart_sampled_baseline"
        },
        "smart_retry_T08": {
          "passed": 62,
          "n": 100,
          "input": 55693,
          "output": 49114,
          "tokens": 104807,
          "inferenceSeconds": 1562.6861354990688,
          "checkSeconds": 168.41779904300347,
          "baseline": "smart_sampled_baseline"
        },
        "oracle_one_resample": {
          "passed": 64,
          "n": 100,
          "input": 55918,
          "output": 61135,
          "tokens": 117053,
          "inferenceSeconds": 1936.2875594568177,
          "checkSeconds": 156.94532253903162,
          "baseline": "baseline"
        },
        "oracle_one_repair": {
          "passed": 66,
          "n": 100,
          "input": 83390,
          "output": 62487,
          "tokens": 145877,
          "inferenceSeconds": 2003.0015907898924,
          "checkSeconds": 139.05761808199168,
          "baseline": "baseline"
        },
        "generated_one_repair": {
          "passed": 57,
          "n": 100,
          "input": 85764,
          "output": 71163,
          "tokens": 156927,
          "inferenceSeconds": 2306.124748002825,
          "checkSeconds": 189.6730120429711,
          "baseline": "baseline"
        },
        "production_fast": {
          "passed": 57,
          "n": 100,
          "input": 85764,
          "output": 71163,
          "tokens": 156927,
          "inferenceSeconds": 2306.124748002825,
          "checkSeconds": 189.6730120429711,
          "baseline": "baseline"
        }
      }
    }
  }
}
