{
  "created_from": [
    "expanded-data.json",
    "doc-batch-20260708-final-summary.json",
    "full-qwen__qwen3.6-27b-max8000-summary.json",
    "x-ai__grok-4.5__strict__20260709T073928Z-c5fdf7a96e.json"
  ],
  "models": [
    {
      "label": "GPT-5.5",
      "model": "openai/gpt-5.5",
      "score": 0.9653535382668413,
      "json": 1.0,
      "action": 0.9811320754716981,
      "false": 0,
      "missed": 0,
      "cost": 2.0262250000000006,
      "http": 0,
      "jsonfail": 0,
      "tokens": 150120,
      "run_id": "20260708T051356Z-ef5567941d",
      "source_group": "doc-batch-20260708",
      "artifact_id": "openai__gpt-5.5__strict__20260708T051356Z-ef5567941d.json",
      "prompt_mode": "strict",
      "repeats": 1,
      "cases": 106,
      "categories": 16
    },
    {
      "label": "Qwen 3.7 Max",
      "model": "qwen/qwen3.7-max",
      "score": 0.964055420076546,
      "json": 1.0,
      "action": 0.9905660377358491,
      "false": 0,
      "missed": 0,
      "cost": 0.71436375,
      "http": 0,
      "jsonfail": 0,
      "tokens": 266903,
      "run_id": "20260708T051356Z-c16015de07",
      "source_group": "doc-batch-20260708",
      "artifact_id": "qwen__qwen3.7-max__strict__20260708T051356Z-c16015de07.json",
      "prompt_mode": "strict",
      "repeats": 1,
      "cases": 106,
      "categories": 16
    },
    {
      "label": "Qwen 3.6 27B",
      "model": "qwen/qwen3.6-27b",
      "score": 0.9620379153000008,
      "json": 1.0,
      "action": 0.9716981132075472,
      "false": 0,
      "missed": 0,
      "cost": 0.76180498,
      "http": 0,
      "jsonfail": 0,
      "tokens": 416720,
      "run_id": "20260708T052944Z-a64e377805",
      "source_group": "doc-batch-20260708-qwen36-max8000",
      "artifact_id": "qwen__qwen3.6-27b__strict__20260708T052944Z-a64e377805.json",
      "prompt_mode": "strict",
      "repeats": 1,
      "cases": 106,
      "categories": 16
    },
    {
      "label": "Grok 4.5",
      "model": "x-ai/grok-4.5",
      "score": 0.9582440047366646,
      "json": 1.0,
      "action": 0.9716981132075472,
      "false": 0,
      "missed": 0,
      "cost": 1.131042,
      "http": 0,
      "jsonfail": 0,
      "tokens": 277753,
      "run_id": "20260709T073928Z-c5fdf7a96e",
      "source_group": "grok-4.5-20260709",
      "artifact_id": "x-ai__grok-4.5__strict__20260709T073928Z-c5fdf7a96e.json",
      "prompt_mode": "strict",
      "repeats": 1,
      "cases": 106,
      "categories": 16
    },
    {
      "label": "GLM 5.2",
      "model": "z-ai/glm-5.2",
      "score": 0.9405168084374604,
      "json": 1.0,
      "action": 0.9339622641509434,
      "false": 0,
      "missed": 5,
      "cost": 0.47946947999999995,
      "http": 0,
      "jsonfail": 0,
      "tokens": 156742,
      "run_id": "20260707T141357Z-6a71127d3f",
      "source_group": "expanded-baselines-20260707",
      "artifact_id": "glm-5.2.json",
      "prompt_mode": "strict",
      "repeats": 1,
      "cases": 106,
      "categories": 16
    },
    {
      "label": "DeepSeek V4 Pro",
      "model": "deepseek/deepseek-v4-pro",
      "score": 0.9402407352799474,
      "json": 0.9905660377358491,
      "action": 0.9339622641509434,
      "false": 0,
      "missed": 6,
      "cost": 0.6135073507999999,
      "http": 1,
      "jsonfail": 1,
      "tokens": 246593,
      "run_id": "20260707T141357Z-3b6fce3999",
      "source_group": "expanded-baselines-20260707",
      "artifact_id": "deepseek-v4-pro.json",
      "prompt_mode": "strict",
      "repeats": 1,
      "cases": 106,
      "categories": 16
    },
    {
      "label": "DeepSeek V4 Flash",
      "model": "deepseek/deepseek-v4-flash",
      "score": 0.9295464679620669,
      "json": 0.9905660377358491,
      "action": 0.9433962264150944,
      "false": 0,
      "missed": 5,
      "cost": 0.03273153319999999,
      "http": 0,
      "jsonfail": 1,
      "tokens": 193634,
      "run_id": "20260707T161318Z-c8f1f5311b",
      "source_group": "expanded-baselines-20260707",
      "artifact_id": "deepseek__deepseek-v4-flash__strict__20260707T161318Z-c8f1f5311b.json",
      "prompt_mode": "strict",
      "repeats": 1,
      "cases": 106,
      "categories": 16
    },
    {
      "label": "MiniMax M3",
      "model": "minimax/minimax-m3",
      "score": 0.9276703028135811,
      "json": 1.0,
      "action": 0.9339622641509434,
      "false": 0,
      "missed": 0,
      "cost": 0.13272833999999997,
      "http": 0,
      "jsonfail": 0,
      "tokens": 202114,
      "run_id": "20260708T051356Z-a89ccb8d9f",
      "source_group": "doc-batch-20260708",
      "artifact_id": "minimax__minimax-m3__strict__20260708T051356Z-a89ccb8d9f.json",
      "prompt_mode": "strict",
      "repeats": 1,
      "cases": 106,
      "categories": 16
    },
    {
      "label": "Grok 4.3",
      "model": "x-ai/grok-4.3",
      "score": 0.9230167553256973,
      "json": 1.0,
      "action": 0.9150943396226415,
      "false": 0,
      "missed": 0,
      "cost": 0.33924289999999985,
      "http": 0,
      "jsonfail": 0,
      "tokens": 200322,
      "run_id": "20260708T051356Z-e2a3724944",
      "source_group": "doc-batch-20260708",
      "artifact_id": "x-ai__grok-4.3__strict__20260708T051356Z-e2a3724944.json",
      "prompt_mode": "strict",
      "repeats": 1,
      "cases": 106,
      "categories": 16
    },
    {
      "label": "Claude Sonnet 5",
      "model": "anthropic/claude-sonnet-5",
      "score": 0.8722267373870466,
      "json": 0.9339622641509434,
      "action": 0.8584905660377359,
      "false": 0,
      "missed": 10,
      "cost": 0.681678,
      "http": 0,
      "jsonfail": 7,
      "tokens": 194319,
      "run_id": "20260707T130202Z-a4347f530c",
      "source_group": "expanded-baselines-20260707",
      "artifact_id": "20260707T130202Z-a4347f530c.json",
      "prompt_mode": "strict",
      "repeats": 1,
      "cases": 106,
      "categories": 16
    },
    {
      "label": "Claude Opus 4.8",
      "model": "anthropic/claude-opus-4.8",
      "score": 0.8370687695658428,
      "json": 0.9905660377358491,
      "action": 0.8679245283018868,
      "false": 0,
      "missed": 10,
      "cost": 1.3728649999999996,
      "http": 0,
      "jsonfail": 1,
      "tokens": 183222,
      "run_id": "20260707T161318Z-08e27cc1bc",
      "source_group": "expanded-baselines-20260707",
      "artifact_id": "anthropic__claude-opus-4.8__strict__20260707T161318Z-08e27cc1bc.json",
      "prompt_mode": "strict",
      "repeats": 1,
      "cases": 106,
      "categories": 16
    },
    {
      "label": "Kimi K2.7 Code",
      "model": "moonshotai/kimi-k2.7-code",
      "score": 0.8026869660521416,
      "json": 0.839622641509434,
      "action": 0.8207547169811321,
      "false": 0,
      "missed": 0,
      "cost": 0.7690549000000003,
      "http": 0,
      "jsonfail": 17,
      "tokens": 272677,
      "run_id": "20260708T051356Z-b1279adf92",
      "source_group": "doc-batch-20260708",
      "artifact_id": "moonshotai__kimi-k2.7-code__strict__20260708T051356Z-b1279adf92.json",
      "prompt_mode": "strict",
      "repeats": 1,
      "cases": 106,
      "categories": 16
    },
    {
      "label": "Ornith 9B Local",
      "model": "local GGUF",
      "score": 0.715304616364238,
      "json": 1.0,
      "action": 0.89937106918239,
      "false": 9,
      "missed": 13,
      "cost": 0.0,
      "http": 0,
      "jsonfail": 0,
      "tokens": 381135,
      "run_id": "20260707T113837Z-c948c50b94",
      "source_group": "expanded-baselines-20260707",
      "artifact_id": "metrics.json",
      "prompt_mode": "strict",
      "repeats": 1,
      "cases": 106,
      "categories": 16
    },
    {
      "label": "HY3 Free",
      "model": "tencent/hy3:free",
      "score": 0.6809514820810285,
      "json": 0.8301886792452831,
      "action": 0.7924528301886793,
      "false": 0,
      "missed": 10,
      "cost": 0.0,
      "http": 6,
      "jsonfail": 18,
      "tokens": 123548,
      "run_id": "20260707T161318Z-9577a732ac",
      "source_group": "expanded-baselines-20260707",
      "artifact_id": "tencent__hy3_free__strict__20260707T161318Z-9577a732ac.json",
      "prompt_mode": "strict",
      "repeats": 1,
      "cases": 106,
      "categories": 16
    },
    {
      "label": "HY3 Paid",
      "model": "tencent/hy3",
      "score": 0.6614551514362905,
      "json": 0.8207547169811321,
      "action": 0.7547169811320755,
      "false": 0,
      "missed": 0,
      "cost": 0.045913999999999996,
      "http": 19,
      "jsonfail": 19,
      "tokens": 120376,
      "run_id": "20260708T051356Z-482cef79dd",
      "source_group": "doc-batch-20260708",
      "artifact_id": "tencent__hy3__strict__20260708T051356Z-482cef79dd.json",
      "prompt_mode": "strict",
      "repeats": 1,
      "cases": 106,
      "categories": 16
    }
  ]
}
