{
  "source_commit": "672227c6b6676edc20d57ea53b7000262aae77b9",
  "checked": "2026-09-07",
  "kind": "Sierra-reported submission metadata; not independently rerun",
  "entries": [
    {
      "model": "GPT-5.6-sol",
      "reasoning_effort": "xhigh",
      "evaluation_date": "2026-07-22",
      "submission_date": "2026-08-04",
      "benchmark_version": "1.0.1",
      "user_simulator": "gpt-5.2",
      "results": {
        "pass_1": 46.90721649484536,
        "pass_2": 37.28522336769759,
        "pass_3": 31.958762886597935,
        "pass_4": 27.835051546391753,
        "cost": null,
        "retrieval_config": "alltools"
      },
      "source": "https://github.com/sierra-research/tau2-bench/blob/672227c6b6676edc20d57ea53b7000262aae77b9/web/leaderboard/public/submissions/gpt-5-6-sol_sierra_2026-08-04/submission.json",
      "sha256": "0394128dc04d58ff77846e404ce588d786988a5c2ffcb69a8af09f19c46c97a4"
    },
    {
      "model": "Claude Opus 5",
      "reasoning_effort": "max",
      "evaluation_date": "2026-08-03",
      "submission_date": "2026-08-04",
      "benchmark_version": "1.0.1",
      "user_simulator": "gpt-5.2",
      "results": {
        "pass_1": 48.71134020618557,
        "pass_2": 39.175257731958766,
        "pass_3": 34.5360824742268,
        "pass_4": 31.958762886597935,
        "cost": null,
        "retrieval_config": "alltools"
      },
      "source": "https://github.com/sierra-research/tau2-bench/blob/672227c6b6676edc20d57ea53b7000262aae77b9/web/leaderboard/public/submissions/claude-opus-5_sierra_2026-08-04/submission.json",
      "sha256": "4ab17f3f2e77cae8b2b4e4ba13413f7e2168104a772f25afe9d071950ed24ba2"
    }
  ]
}
