{
  "v1_1": {
    "models": [
      "GPT-5.4-mini",
      "Claude Sonnet 4.6",
      "Claude Fable 5",
      "Claude Opus 4.6",
      "GPT-5.5",
      "Claude Opus 4.7",
      "Claude Opus 4.8",
      "Claude Sonnet 5"
    ],
    "colors": {
      "GPT-5.4-mini": "#10a37f",
      "Claude Sonnet 4.6": "#d97757",
      "Claude Fable 5": "#d97757",
      "Claude Opus 4.6": "#d97757",
      "GPT-5.5": "#10a37f",
      "Claude Opus 4.7": "#d97757",
      "Claude Opus 4.8": "#d97757",
      "Claude Sonnet 5": "#d97757"
    },
    "lab_colors": {
      "OpenAI": "#10a37f",
      "Anthropic": "#d97757"
    },
    "harness": {
      "GPT-5.4-mini": "codex",
      "Claude Sonnet 4.6": "claude-code",
      "Claude Fable 5": "claude-code",
      "Claude Opus 4.6": "claude-code",
      "GPT-5.5": "codex",
      "Claude Opus 4.7": "claude-code",
      "Claude Opus 4.8": "claude-code",
      "Claude Sonnet 5": "claude-code"
    },
    "efforts": {
      "GPT-5.4-mini": ["low", "medium", "high", "xhigh"],
      "Claude Sonnet 4.6": ["low", "medium", "high", "max"],
      "Claude Fable 5": ["low", "medium", "high", "xhigh", "max"],
      "Claude Opus 4.6": ["low", "medium", "high", "max"],
      "GPT-5.5": ["low", "medium", "high", "xhigh"],
      "Claude Opus 4.7": ["low", "medium", "high", "xhigh", "max"],
      "Claude Opus 4.8": ["low", "medium", "high", "xhigh", "max"],
      "Claude Sonnet 5": ["low", "medium", "high", "xhigh", "max"]
    },
    "subsets": {
      "main": 100,
      "extended": 150
    },
    "data": {
      "GPT-5.4-mini": {
        "low": {
          "main": {
            "correct": 0.106,
            "new_score": 0.0925,
            "tokens": 12065.67,
            "cost": 0.3642,
            "duration_min": 3.25,
            "tool_calls": 43.69,
            "steps": null,
            "ote": 80932.5,
            "shortcut_rate_1_0": 0.0071,
            "tokens_1_0": 11463.82
          },
          "extended": {
            "correct": 0.2302,
            "new_score": 0.2058,
            "tokens": 10771.09,
            "cost": 0.3288,
            "duration_min": 2.96,
            "tool_calls": 40.32,
            "steps": null,
            "ote": 73066.5,
            "shortcut_rate_1_0": 0.0047,
            "tokens_1_0": 10242.59
          }
        },
        "medium": {
          "main": {
            "correct": 0.2373,
            "new_score": 0.2078,
            "tokens": 34721.25,
            "cost": 0.7833,
            "duration_min": 6.33,
            "tool_calls": 76.33,
            "steps": null,
            "ote": 174059.4,
            "shortcut_rate_1_0": 0.0162,
            "tokens_1_0": 35064.1
          },
          "extended": {
            "correct": 0.3922,
            "new_score": 0.3506,
            "tokens": 29564.36,
            "cost": 0.6716,
            "duration_min": 5.5,
            "tool_calls": 68.05,
            "steps": null,
            "ote": 149239.5,
            "shortcut_rate_1_0": 0.0141,
            "tokens_1_0": 29894.07
          }
        },
        "high": {
          "main": {
            "correct": 0.263,
            "new_score": 0.2308,
            "tokens": 46438.82,
            "cost": 0.9537,
            "duration_min": 8.25,
            "tool_calls": 88.1,
            "steps": null,
            "ote": 211928.3,
            "shortcut_rate_1_0": 0.0384,
            "tokens_1_0": 45289.64
          },
          "extended": {
            "correct": 0.4153,
            "new_score": 0.3719,
            "tokens": 39150.98,
            "cost": 0.8043,
            "duration_min": 6.94,
            "tool_calls": 77.86,
            "steps": null,
            "ote": 178728.7,
            "shortcut_rate_1_0": 0.0302,
            "tokens_1_0": 38178.51
          }
        },
        "xhigh": {
          "main": {
            "correct": 0.308,
            "new_score": 0.2704,
            "tokens": 90933.29,
            "cost": 1.5201,
            "duration_min": 13.56,
            "tool_calls": 135.41,
            "steps": null,
            "ote": 337806.7,
            "shortcut_rate_1_0": 0.0758,
            "tokens_1_0": 88876.4
          },
          "extended": {
            "correct": 0.4794,
            "new_score": 0.4301,
            "tokens": 75608.73,
            "cost": 1.2621,
            "duration_min": 11.23,
            "tool_calls": 115.92,
            "steps": null,
            "ote": 280457.1,
            "shortcut_rate_1_0": 0.0604,
            "tokens_1_0": 73806.65
          }
        }
      },
      "Claude Sonnet 4.6": {
        "low": {
          "main": {
            "correct": 0.1691,
            "new_score": 0.1503,
            "tokens": 21707.59,
            "cost": 1.6481,
            "duration_min": 9.91,
            "tool_calls": 86.82,
            "steps": null,
            "ote": 122159.4,
            "shortcut_rate_1_0": 0.0051,
            "tokens_1_0": 18548.78
          },
          "extended": {
            "correct": 0.3365,
            "new_score": 0.3044,
            "tokens": 18901.69,
            "cost": 1.4261,
            "duration_min": 8.73,
            "tool_calls": 77.04,
            "steps": null,
            "ote": 106031.1,
            "shortcut_rate_1_0": 0.0074,
            "tokens_1_0": 15820.72
          }
        },
        "medium": {
          "main": {
            "correct": 0.239,
            "new_score": 0.211,
            "tokens": 33357.25,
            "cost": 2.3012,
            "duration_min": 13.18,
            "tool_calls": 101.34,
            "steps": null,
            "ote": 168739,
            "shortcut_rate_1_0": 0.0061,
            "tokens_1_0": 27711.01
          },
          "extended": {
            "correct": 0.4167,
            "new_score": 0.3749,
            "tokens": 28942.96,
            "cost": 1.9375,
            "duration_min": 11.47,
            "tool_calls": 89.03,
            "steps": null,
            "ote": 143502.2,
            "shortcut_rate_1_0": 0.0047,
            "tokens_1_0": 23903.87
          }
        },
        "high": {
          "main": {
            "correct": 0.265,
            "new_score": 0.2347,
            "tokens": 41108.89,
            "cost": 2.6993,
            "duration_min": 15.78,
            "tool_calls": 107.08,
            "steps": null,
            "ote": 194996.1,
            "shortcut_rate_1_0": 0.0081,
            "tokens_1_0": 34413.31
          },
          "extended": {
            "correct": 0.4412,
            "new_score": 0.3986,
            "tokens": 35510.39,
            "cost": 2.2694,
            "duration_min": 13.64,
            "tool_calls": 94.04,
            "steps": null,
            "ote": 165538.5,
            "shortcut_rate_1_0": 0.0067,
            "tokens_1_0": 29774.25
          }
        },
        "max": {
          "main": {
            "correct": 0.275,
            "new_score": 0.2431,
            "tokens": 44191.74,
            "cost": 2.8979,
            "duration_min": 16.53,
            "tool_calls": 111.87,
            "steps": null,
            "ote": 204976.6,
            "shortcut_rate_1_0": 0.0071,
            "tokens_1_0": 37588.94
          },
          "extended": {
            "correct": 0.444,
            "new_score": 0.4,
            "tokens": 38859.68,
            "cost": 2.4401,
            "duration_min": 14.45,
            "tool_calls": 97.78,
            "steps": null,
            "ote": 174589.4,
            "shortcut_rate_1_0": 0.0047,
            "tokens_1_0": 32700.86
          }
        }
      },
      "Claude Fable 5": {
        "low": {
          "main": {
            "correct": 0.5282,
            "new_score": 0.4801,
            "tokens": 23609.82,
            "cost": 4.9232,
            "duration_min": 15.19,
            "tool_calls": 50.77,
            "steps": null,
            "ote": 98405.8,
            "shortcut_rate_1_0": 0.101,
            "tokens_1_0": 21262.22
          },
          "extended": {
            "correct": 0.663,
            "new_score": 0.608,
            "tokens": 19502.26,
            "cost": 4.0311,
            "duration_min": 12.75,
            "tool_calls": 43.76,
            "steps": null,
            "ote": 80603.4,
            "shortcut_rate_1_0": 0.0805,
            "tokens_1_0": 17747.44
          }
        },
        "medium": {
          "main": {
            "correct": 0.546,
            "new_score": 0.4982,
            "tokens": 33229.87,
            "cost": 7.1539,
            "duration_min": 17.59,
            "tool_calls": 62.09,
            "steps": null,
            "ote": 142017.3,
            "shortcut_rate_1_0": 0.2141,
            "tokens_1_0": 27155.51
          },
          "extended": {
            "correct": 0.6825,
            "new_score": 0.6279,
            "tokens": 27536.09,
            "cost": 5.8428,
            "duration_min": 15.02,
            "tool_calls": 53.47,
            "steps": null,
            "ote": 116151.3,
            "shortcut_rate_1_0": 0.1718,
            "tokens_1_0": 23104.14
          }
        },
        "high": {
          "main": {
            "correct": 0.5766,
            "new_score": 0.5273,
            "tokens": 43482.76,
            "cost": 9.4805,
            "duration_min": 19.74,
            "tool_calls": 73.3,
            "steps": null,
            "ote": 189376.6,
            "shortcut_rate_1_0": 0.3111,
            "tokens_1_0": 34776.93
          },
          "extended": {
            "correct": 0.6981,
            "new_score": 0.6425,
            "tokens": 36012.96,
            "cost": 7.711,
            "duration_min": 16.81,
            "tool_calls": 62.9,
            "steps": null,
            "ote": 154083.8,
            "shortcut_rate_1_0": 0.2349,
            "tokens_1_0": 29252.91
          }
        },
        "xhigh": {
          "main": {
            "correct": 0.5885,
            "new_score": 0.5348,
            "tokens": 58557.89,
            "cost": 13.0938,
            "duration_min": 23.76,
            "tool_calls": 89.14,
            "steps": null,
            "ote": 261896.2,
            "shortcut_rate_1_0": 0.3717,
            "tokens_1_0": 45395.57
          },
          "extended": {
            "correct": 0.7086,
            "new_score": 0.6494,
            "tokens": 48419.27,
            "cost": 10.5282,
            "duration_min": 19.98,
            "tool_calls": 75.65,
            "steps": null,
            "ote": 210591.7,
            "shortcut_rate_1_0": 0.3074,
            "tokens_1_0": 38379.85
          }
        },
        "max": {
          "main": {
            "correct": 0.5689,
            "new_score": 0.5157,
            "tokens": 80761.13,
            "cost": 19.0696,
            "duration_min": 29.5,
            "tool_calls": 107.34,
            "steps": null,
            "ote": 377912,
            "shortcut_rate_1_0": 0.3798,
            "tokens_1_0": 65654.18
          },
          "extended": {
            "correct": 0.6956,
            "new_score": 0.6363,
            "tokens": 68418.61,
            "cost": 15.4885,
            "duration_min": 25.31,
            "tool_calls": 91.95,
            "steps": null,
            "ote": 307494,
            "shortcut_rate_1_0": 0.3248,
            "tokens_1_0": 56060.3
          }
        }
      },
      "Claude Opus 4.6": {
        "low": {
          "main": {
            "correct": 0.2054,
            "new_score": 0.1831,
            "tokens": 14888.41,
            "cost": 2.101,
            "duration_min": 8.55,
            "tool_calls": 77.7,
            "steps": null,
            "ote": 94991.6,
            "shortcut_rate_1_0": 0.0061,
            "tokens_1_0": 12301.04
          },
          "extended": {
            "correct": 0.3643,
            "new_score": 0.3293,
            "tokens": 13032.03,
            "cost": 1.8406,
            "duration_min": 7.68,
            "tool_calls": 67.46,
            "steps": null,
            "ote": 82588.9,
            "shortcut_rate_1_0": 0.0087,
            "tokens_1_0": 10813.48
          }
        },
        "medium": {
          "main": {
            "correct": 0.265,
            "new_score": 0.2367,
            "tokens": 19836.5,
            "cost": 2.7941,
            "duration_min": 10.43,
            "tool_calls": 95.63,
            "steps": null,
            "ote": 126589.1,
            "shortcut_rate_1_0": 0.0051,
            "tokens_1_0": 16304.72
          },
          "extended": {
            "correct": 0.426,
            "new_score": 0.3865,
            "tokens": 17306.79,
            "cost": 2.3912,
            "duration_min": 9.16,
            "tool_calls": 82.52,
            "steps": null,
            "ote": 108478.8,
            "shortcut_rate_1_0": 0.0081,
            "tokens_1_0": 14167.74
          }
        },
        "high": {
          "main": {
            "correct": 0.297,
            "new_score": 0.2664,
            "tokens": 26676.82,
            "cost": 3.9796,
            "duration_min": 13.36,
            "tool_calls": 120.46,
            "steps": null,
            "ote": 168409.1,
            "shortcut_rate_1_0": 0.0071,
            "tokens_1_0": 21822.24
          },
          "extended": {
            "correct": 0.48,
            "new_score": 0.4367,
            "tokens": 23162.12,
            "cost": 3.2978,
            "duration_min": 11.65,
            "tool_calls": 102.2,
            "steps": null,
            "ote": 142176.8,
            "shortcut_rate_1_0": 0.0067,
            "tokens_1_0": 18906.2
          }
        },
        "max": {
          "main": {
            "correct": 0.303,
            "new_score": 0.269,
            "tokens": 28870.01,
            "cost": 3.7346,
            "duration_min": 14.06,
            "tool_calls": 114.32,
            "steps": null,
            "ote": 165028.1,
            "shortcut_rate_1_0": 0.0051,
            "tokens_1_0": 23359.61
          },
          "extended": {
            "correct": 0.4738,
            "new_score": 0.429,
            "tokens": 24940.95,
            "cost": 3.1206,
            "duration_min": 12.18,
            "tool_calls": 98.75,
            "steps": null,
            "ote": 140032.8,
            "shortcut_rate_1_0": 0.0067,
            "tokens_1_0": 20387.21
          }
        }
      },
      "GPT-5.5": {
        "low": {
          "main": {
            "correct": 0.3439,
            "new_score": 0.3055,
            "tokens": 8854.97,
            "cost": 1.6478,
            "duration_min": 5.3,
            "tool_calls": 39.82,
            "steps": null,
            "ote": 54926.9,
            "shortcut_rate_1_0": 0.002,
            "tokens_1_0": 8623.17
          },
          "extended": {
            "correct": 0.5039,
            "new_score": 0.4555,
            "tokens": 7726.62,
            "cost": 1.434,
            "duration_min": 4.65,
            "tool_calls": 35.78,
            "steps": null,
            "ote": 47798.6,
            "shortcut_rate_1_0": 0.002,
            "tokens_1_0": 7493.67
          }
        },
        "medium": {
          "main": {
            "correct": 0.4108,
            "new_score": 0.366,
            "tokens": 13677.18,
            "cost": 2.3513,
            "duration_min": 6.87,
            "tool_calls": 55.78,
            "steps": null,
            "ote": 78376.7,
            "shortcut_rate_1_0": 0.0202,
            "tokens_1_0": 13662.86
          },
          "extended": {
            "correct": 0.5679,
            "new_score": 0.5137,
            "tokens": 11724.29,
            "cost": 1.9847,
            "duration_min": 5.91,
            "tool_calls": 49.24,
            "steps": null,
            "ote": 66157.7,
            "shortcut_rate_1_0": 0.0168,
            "tokens_1_0": 11695.41
          }
        },
        "high": {
          "main": {
            "correct": 0.4571,
            "new_score": 0.4056,
            "tokens": 18422.13,
            "cost": 3.1187,
            "duration_min": 8.54,
            "tool_calls": 68.29,
            "steps": null,
            "ote": 103958.3,
            "shortcut_rate_1_0": 0.0283,
            "tokens_1_0": 18621.15
          },
          "extended": {
            "correct": 0.5987,
            "new_score": 0.5399,
            "tokens": 15774.98,
            "cost": 2.6054,
            "duration_min": 7.29,
            "tool_calls": 60.02,
            "steps": null,
            "ote": 86845.8,
            "shortcut_rate_1_0": 0.0255,
            "tokens_1_0": 15917.14
          }
        },
        "xhigh": {
          "main": {
            "correct": 0.4824,
            "new_score": 0.4296,
            "tokens": 24992.38,
            "cost": 4.0348,
            "duration_min": 10.3,
            "tool_calls": 84.28,
            "steps": null,
            "ote": 134494.1,
            "shortcut_rate_1_0": 0.0384,
            "tokens_1_0": 24903.27
          },
          "extended": {
            "correct": 0.6276,
            "new_score": 0.5666,
            "tokens": 21330.94,
            "cost": 3.3393,
            "duration_min": 8.77,
            "tool_calls": 73.4,
            "steps": null,
            "ote": 111311.6,
            "shortcut_rate_1_0": 0.0329,
            "tokens_1_0": 21257.25
          }
        }
      },
      "Claude Opus 4.7": {
        "low": {
          "main": {
            "correct": 0.3094,
            "new_score": 0.2761,
            "tokens": 17852.73,
            "cost": 2.4668,
            "duration_min": 8.26,
            "tool_calls": 53.37,
            "steps": null,
            "ote": 101521.8,
            "shortcut_rate_1_0": 0.0253,
            "tokens_1_0": 16155.8
          },
          "extended": {
            "correct": 0.4783,
            "new_score": 0.436,
            "tokens": 15368.95,
            "cost": 2.1079,
            "duration_min": 7.28,
            "tool_calls": 46.69,
            "steps": null,
            "ote": 86734.8,
            "shortcut_rate_1_0": 0.0221,
            "tokens_1_0": 13864.7
          }
        },
        "medium": {
          "main": {
            "correct": 0.3223,
            "new_score": 0.2891,
            "tokens": 21880.82,
            "cost": 3.2194,
            "duration_min": 9.77,
            "tool_calls": 63.95,
            "steps": null,
            "ote": 131993.1,
            "shortcut_rate_1_0": 0.0333,
            "tokens_1_0": 20038.73
          },
          "extended": {
            "correct": 0.4902,
            "new_score": 0.4475,
            "tokens": 18778.34,
            "cost": 2.7316,
            "duration_min": 8.52,
            "tool_calls": 55.74,
            "steps": null,
            "ote": 112064.5,
            "shortcut_rate_1_0": 0.0302,
            "tokens_1_0": 16946.15
          }
        },
        "high": {
          "main": {
            "correct": 0.3924,
            "new_score": 0.3517,
            "tokens": 30140.22,
            "cost": 5.0292,
            "duration_min": 13.16,
            "tool_calls": 84.22,
            "steps": null,
            "ote": 202824,
            "shortcut_rate_1_0": 0.0354,
            "tokens_1_0": 27778.02
          },
          "extended": {
            "correct": 0.5608,
            "new_score": 0.5112,
            "tokens": 25742.05,
            "cost": 4.2117,
            "duration_min": 11.24,
            "tool_calls": 72.7,
            "steps": null,
            "ote": 170038.5,
            "shortcut_rate_1_0": 0.0369,
            "tokens_1_0": 23354.2
          }
        },
        "xhigh": {
          "main": {
            "correct": 0.3867,
            "new_score": 0.3491,
            "tokens": 38418.65,
            "cost": 6.8688,
            "duration_min": 15.48,
            "tool_calls": 103.92,
            "steps": null,
            "ote": 276340.3,
            "shortcut_rate_1_0": 0.0414,
            "tokens_1_0": 34294.3
          },
          "extended": {
            "correct": 0.5618,
            "new_score": 0.5141,
            "tokens": 32928.56,
            "cost": 5.7683,
            "duration_min": 13.24,
            "tool_calls": 90.4,
            "steps": null,
            "ote": 232185.7,
            "shortcut_rate_1_0": 0.0376,
            "tokens_1_0": 29274.29
          }
        },
        "max": {
          "main": {
            "correct": 0.4283,
            "new_score": 0.3854,
            "tokens": 49113.7,
            "cost": 9.0876,
            "duration_min": 19.83,
            "tool_calls": 126.29,
            "steps": null,
            "ote": 365273.2,
            "shortcut_rate_1_0": 0.0404,
            "tokens_1_0": 43092.06
          },
          "extended": {
            "correct": 0.5909,
            "new_score": 0.5393,
            "tokens": 42112.91,
            "cost": 7.6144,
            "duration_min": 17.05,
            "tool_calls": 110.69,
            "steps": null,
            "ote": 306462.3,
            "shortcut_rate_1_0": 0.0416,
            "tokens_1_0": 36919.01
          }
        }
      },
      "Claude Opus 4.8": {
        "low": {
          "main": {
            "correct": 0.3923,
            "new_score": 0.3514,
            "tokens": 25194.01,
            "cost": 2.6827,
            "duration_min": 10.81,
            "tool_calls": 66.84,
            "steps": null,
            "ote": 116691.8,
            "shortcut_rate_1_0": 0.0384,
            "tokens_1_0": 22399.15
          },
          "extended": {
            "correct": 0.5601,
            "new_score": 0.5083,
            "tokens": 21383.56,
            "cost": 2.2736,
            "duration_min": 9.35,
            "tool_calls": 58.15,
            "steps": null,
            "ote": 98911.4,
            "shortcut_rate_1_0": 0.0315,
            "tokens_1_0": 18855.64
          }
        },
        "medium": {
          "main": {
            "correct": 0.4493,
            "new_score": 0.4046,
            "tokens": 34906.51,
            "cost": 3.7029,
            "duration_min": 13.46,
            "tool_calls": 76.43,
            "steps": null,
            "ote": 157592.4,
            "shortcut_rate_1_0": 0.0444,
            "tokens_1_0": 30462.94
          },
          "extended": {
            "correct": 0.6076,
            "new_score": 0.5537,
            "tokens": 29301.99,
            "cost": 3.062,
            "duration_min": 11.56,
            "tool_calls": 66.55,
            "steps": null,
            "ote": 130736.2,
            "shortcut_rate_1_0": 0.0369,
            "tokens_1_0": 25499.96
          }
        },
        "high": {
          "main": {
            "correct": 0.4565,
            "new_score": 0.4102,
            "tokens": 41326.36,
            "cost": 4.2799,
            "duration_min": 14.86,
            "tool_calls": 79.36,
            "steps": null,
            "ote": 180048.3,
            "shortcut_rate_1_0": 0.0566,
            "tokens_1_0": 36780.65
          },
          "extended": {
            "correct": 0.611,
            "new_score": 0.5563,
            "tokens": 34584.21,
            "cost": 3.5377,
            "duration_min": 12.82,
            "tool_calls": 69,
            "steps": null,
            "ote": 149024.5,
            "shortcut_rate_1_0": 0.0443,
            "tokens_1_0": 30736.38
          }
        },
        "xhigh": {
          "main": {
            "correct": 0.5027,
            "new_score": 0.4549,
            "tokens": 65742.42,
            "cost": 6.7803,
            "duration_min": 19.72,
            "tool_calls": 93.56,
            "steps": null,
            "ote": 276888.8,
            "shortcut_rate_1_0": 0.0768,
            "tokens_1_0": 60385.53
          },
          "extended": {
            "correct": 0.6504,
            "new_score": 0.5945,
            "tokens": 55453.86,
            "cost": 5.5793,
            "duration_min": 16.97,
            "tool_calls": 81.15,
            "steps": null,
            "ote": 227811.8,
            "shortcut_rate_1_0": 0.0664,
            "tokens_1_0": 50587.62
          }
        },
        "max": {
          "main": {
            "correct": 0.5165,
            "new_score": 0.465,
            "tokens": 95939.85,
            "cost": 9.6225,
            "duration_min": 25.99,
            "tool_calls": 111.64,
            "steps": null,
            "ote": 392048.3,
            "shortcut_rate_1_0": 0.0939,
            "tokens_1_0": 87657.99
          },
          "extended": {
            "correct": 0.6549,
            "new_score": 0.596,
            "tokens": 82896.6,
            "cost": 8.0501,
            "duration_min": 22.63,
            "tool_calls": 97.62,
            "steps": null,
            "ote": 327931.4,
            "shortcut_rate_1_0": 0.0846,
            "tokens_1_0": 75301.11
          }
        }
      },
      "Claude Sonnet 5": {
        "low": {
          "main": {
            "correct": 0.32,
            "new_score": 0.2871,
            "tokens": 24817.11,
            "cost": 2.1507,
            "duration_min": 13.8,
            "tool_calls": 91.82,
            "steps": null,
            "ote": 154483.7,
            "shortcut_rate_1_0": 0.0303,
            "tokens_1_0": 11219.45
          },
          "extended": {
            "correct": 0.4891,
            "new_score": 0.4446,
            "tokens": 21669.84,
            "cost": 1.8305,
            "duration_min": 12.38,
            "tool_calls": 79.45,
            "steps": null,
            "ote": 130545.4,
            "shortcut_rate_1_0": 0.0215,
            "tokens_1_0": 10084.09
          }
        },
        "medium": {
          "main": {
            "correct": 0.3907,
            "new_score": 0.352,
            "tokens": 38268.93,
            "cost": 3.4334,
            "duration_min": 17.73,
            "tool_calls": 106.57,
            "steps": null,
            "ote": 236940.2,
            "shortcut_rate_1_0": 0.0808,
            "tokens_1_0": 32178.93
          },
          "extended": {
            "correct": 0.5413,
            "new_score": 0.4928,
            "tokens": 32169.19,
            "cost": 2.8424,
            "duration_min": 15.54,
            "tool_calls": 91.91,
            "steps": null,
            "ote": 196024.7,
            "shortcut_rate_1_0": 0.0597,
            "tokens_1_0": 26884.76
          }
        },
        "high": {
          "main": {
            "correct": 0.4366,
            "new_score": 0.3935,
            "tokens": 57709.84,
            "cost": 5.4879,
            "duration_min": 22.89,
            "tool_calls": 130.38,
            "steps": null,
            "ote": 372886.6,
            "shortcut_rate_1_0": 0.1576,
            "tokens_1_0": 49353.18
          },
          "extended": {
            "correct": 0.5784,
            "new_score": 0.5271,
            "tokens": 48028.6,
            "cost": 4.4766,
            "duration_min": 19.53,
            "tool_calls": 112.23,
            "steps": null,
            "ote": 303880.1,
            "shortcut_rate_1_0": 0.1295,
            "tokens_1_0": 40546.83
          }
        },
        "xhigh": {
          "main": {
            "correct": 0.4759,
            "new_score": 0.4273,
            "tokens": 87937.47,
            "cost": 9.0599,
            "duration_min": 30.82,
            "tool_calls": 164.78,
            "steps": null,
            "ote": 600536.4,
            "shortcut_rate_1_0": 0.2616,
            "tokens_1_0": 74075.29
          },
          "extended": {
            "correct": 0.6166,
            "new_score": 0.5618,
            "tokens": 73988.36,
            "cost": 7.4079,
            "duration_min": 26.26,
            "tool_calls": 142.19,
            "steps": null,
            "ote": 490978.8,
            "shortcut_rate_1_0": 0.2208,
            "tokens_1_0": 61290.36
          }
        },
        "max": {
          "main": {
            "correct": 0.4719,
            "new_score": 0.4236,
            "tokens": 142620.64,
            "cost": 15.4048,
            "duration_min": 45.97,
            "tool_calls": 226.92,
            "steps": null,
            "ote": 1010353.2,
            "shortcut_rate_1_0": 0.3646,
            "tokens_1_0": 116209.26
          },
          "extended": {
            "correct": 0.6019,
            "new_score": 0.5485,
            "tokens": 121728.09,
            "cost": 12.6849,
            "duration_min": 39.66,
            "tool_calls": 196.23,
            "steps": null,
            "ote": 830840.3,
            "shortcut_rate_1_0": 0.3174,
            "tokens_1_0": 99455.81
          }
        }
      }
    }
  }
}
