{
  "run_id": "2026-10-08-core-greedy-nothink-ee86f5",
  "suite": "decision-conformance",
  "tier": "core",
  "suite_hash": "554e796a858218e48bcb83da003554769782a4a32d3e061da87510d7f9cea490",
  "subject": "d1-omni-600m/hf-02b55d7/fp16-liquidai/decide-server-d1-torch-cuda-tf5.19.0-torch2.14.1-cu126",
  "mode": "native",
  "profile": "greedy-nothink",
  "status": "complete",
  "label": null,
  "items": {
    "planned": 17,
    "attempted": 17,
    "succeeded": 17,
    "failed": 0,
    "failed_by_status": {}
  },
  "metrics": {
    "checks": {
      "invariants": {
        "passed": 33,
        "total": 33,
        "failures": []
      },
      "batched_vs_separate": {
        "passed": 6,
        "total": 6,
        "failures": [],
        "max_abs_diff": 0.000055
      },
      "repeat_stability": {
        "passed": 6,
        "total": 6,
        "failures": [],
        "max_std": 0
      },
      "option_order": {
        "passed": 2,
        "total": 2,
        "failures": [],
        "max_confidence_delta": 0.012834
      },
      "monotonic": {
        "passed": 1,
        "total": 2,
        "failures": [
          "state 1 -> 2: 1.051018738609855 -> 0.2864902683068067"
        ],
        "expected_scores": [
          0.626453,
          1.051019,
          0.28649
        ]
      }
    },
    "passed": 48,
    "total": 49
  },
  "latency_ms": {
    "p50": 73.463875,
    "p95": 193.233417,
    "p99": 193.233417,
    "note": "successful requests, nearest rank, measured from America/Chicago"
  },
  "usage": {
    "input_tokens": 4685,
    "output_tokens": 0,
    "cost_usd": 0
  }
}