{
  "schema": "chi-bench/submission/v1",
  "submission": {
    "id": "fable-5-claude-code",
    "team": "Actava",
    "contact": "dark.savi@gmail.com",
    "agent": "claude-code",
    "model": "anthropic/claude-fable-5",
    "notes": "Claude Fable 5 via the stock claude-code harness (MCP tools), routed through OpenRouter.\n",
    "submitted_at": "2026-07-22T23:50:25Z"
  },
  "dataset": {
    "version": "chi-bench-v1.0.0",
    "domains": [
      "pa_provider",
      "pa_um",
      "cm"
    ],
    "name": "chi-bench"
  },
  "results": {
    "overall": {
      "n_trials": 75,
      "n_tasks": 75,
      "pass_at_1": 0.24,
      "mean_cost_usd": 7.0079016,
      "mean_walltime_s": 514.00465704
    },
    "per_domain": {
      "pa_provider": {
        "n_trials": 25,
        "n_tasks": 25,
        "pass_at_1": 0.24,
        "mean_cost_usd": 5.90429888,
        "mean_walltime_s": 339.65884716
      },
      "pa_um": {
        "n_trials": 25,
        "n_tasks": 25,
        "pass_at_1": 0.24,
        "mean_cost_usd": 7.0953063599999995,
        "mean_walltime_s": 482.04023800000004
      },
      "cm": {
        "n_trials": 25,
        "n_tasks": 25,
        "pass_at_1": 0.24,
        "mean_cost_usd": 8.02409956,
        "mean_walltime_s": 720.31488596
      }
    },
    "mean_cost_usd": 7.0079016,
    "mean_walltime_s": 514.00465704
  },
  "provenance": {
    "chi_bench_git_sha": "5b5be35d34ff06f77a2626fb20b096ee653c3962",
    "image_digest": null,
    "judge_model": "claude-opus-4-7",
    "harness_version": "0.1.0",
    "code_dirty": true,
    "dataset_version": "chi-bench-v1.0.0",
    "environment": "modal",
    "started_at": "2026-07-22T07:16:27.207645Z",
    "finished_at": "2026-07-22T10:25:29.658582Z",
    "source": "fable5_openrouter_full_2026_07 experiment matrix"
  }
}