{
  "benchmark": "EPICODE-Bench Mini",
  "version": "1.1.0",
  "case_count": 20,
  "generated_at": "2026-07-18T12:53:15.669599+00:00",
  "prompt_sha256": "d8adf5d0ef83c117369efbdf155d87d136abdc3d149120df9511fddd182c549e",
  "models": [
    "gpt-5.6-sol",
    "gpt-5.6-luna"
  ],
  "runs": [
    {
      "model": "gpt-5.6-sol",
      "reasoning_effort": "xhigh",
      "summary": {
        "resolved": 14,
        "total": 20,
        "resolved_percent": 70.0,
        "code_micro_f1": 100.0,
        "evidence_micro_f1": 88.1,
        "hallucination_percent": 0.0,
        "format_success_percent": 100.0,
        "average_latency_ms": 3923,
        "input_tokens": 14027,
        "output_tokens": 2680,
        "standard_resolved_percent": 40.0,
        "abstention_trap_resolved_percent": 100.0
      }
    },
    {
      "model": "gpt-5.6-luna",
      "reasoning_effort": "low",
      "summary": {
        "resolved": 12,
        "total": 20,
        "resolved_percent": 60.0,
        "code_micro_f1": 83.3,
        "evidence_micro_f1": 71.0,
        "hallucination_percent": 16.7,
        "format_success_percent": 100.0,
        "average_latency_ms": 2366,
        "input_tokens": 14027,
        "output_tokens": 2796,
        "standard_resolved_percent": 20.0,
        "abstention_trap_resolved_percent": 100.0
      }
    }
  ]
}