{
  "url": "https://www.aibattle.space/rankings/eval-briefcase-mean-turns-per-task",
  "dataset": "Mean Turns per Task",
  "family": "tab::eval-briefcase",
  "familyLabel": "eval briefcase",
  "sourceRowCount": 20,
  "window": 20,
  "metricKeys": [
    "aaBriefcaseMeanTurnsPerTask"
  ],
  "rebuilt": "2026-09-19",
  "stats": {
    "median": 99.34615384615384,
    "mean": 102.76310577356844,
    "min": 67.06593406593407,
    "max": 241.98823529411766,
    "spread": 174.92230122818358,
    "gapFirstToSecond": 95.47138136153339,
    "gapFirstToSecondPercent": 65.16,
    "creators": 11
  },
  "entries": [
    {
      "rank": 1,
      "name": "Qwen3.8-Flash-Next",
      "creator": "Alibaba",
      "slug": "qwen3-8-flash-next",
      "detailsUrl": "/models/qwen3-8-flash-next",
      "value": 241.98823529411766
    },
    {
      "rank": 2,
      "name": "GLM-5.3-Flash",
      "creator": "Z AI",
      "slug": "glm-5-3-flash",
      "detailsUrl": "/models/glm-5-3-flash",
      "value": 146.51685393258427
    },
    {
      "rank": 3,
      "name": "gpt-oss-120b (high)",
      "creator": "OpenAI",
      "slug": "gpt-oss-120b",
      "detailsUrl": "/models/gpt-oss-120b",
      "value": 112.71428571428571
    },
    {
      "rank": 4,
      "name": "GLM-5.3 (max)",
      "creator": "Z AI",
      "slug": "glm-5-3",
      "detailsUrl": "/models/glm-5-3",
      "value": 109.85393258426966
    },
    {
      "rank": 5,
      "name": "DeepSeek V4.1 Flash (Reasoning, Max Effort)",
      "creator": "DeepSeek",
      "slug": "deepseek-v4-1-flash",
      "detailsUrl": "/models/deepseek-v4-1-flash",
      "value": 108.91111111111111
    },
    {
      "rank": 6,
      "name": "Claude Fable 5.1 (Adaptive Reasoning, Max Effort, Default Fallback)",
      "creator": "Anthropic",
      "slug": "claude-fable-5-1",
      "detailsUrl": "/models/claude-fable-5-1",
      "value": 105.0989010989011
    },
    {
      "rank": 7,
      "name": "Claude Opus 5 (Adaptive Reasoning, Max Effort)",
      "creator": "Anthropic",
      "slug": "claude-opus-5",
      "detailsUrl": "/models/claude-opus-5",
      "value": 102.73626373626374
    },
    {
      "rank": 8,
      "name": "Qwen3.8 2.4T A95B",
      "creator": "Alibaba",
      "slug": "qwen3-8-2-4t-a95b",
      "detailsUrl": "/models/qwen3-8-2-4t-a95b",
      "value": 100.74725274725274
    },
    {
      "rank": 9,
      "name": "MiniMax-M3",
      "creator": "MiniMax",
      "slug": "minimax-m3",
      "detailsUrl": "/models/minimax-m3",
      "value": 100.65934065934066
    },
    {
      "rank": 10,
      "name": "Gemini 3.8 Flash (high)",
      "creator": "Google",
      "slug": "gemini-3-8-flash",
      "detailsUrl": "/models/gemini-3-8-flash",
      "value": 99.58241758241758
    },
    {
      "rank": 11,
      "name": "Muse Spark 1.3 (max)",
      "creator": "Meta",
      "slug": "muse-spark-1-3",
      "detailsUrl": "/models/muse-spark-1-3",
      "value": 99.10989010989012
    },
    {
      "rank": 12,
      "name": "Claude Opus 5 (Adaptive Reasoning, Xhigh Effort)",
      "creator": "Anthropic",
      "slug": "claude-opus-5-xhigh",
      "detailsUrl": "/models/claude-opus-5-xhigh",
      "value": 90.85714285714286
    },
    {
      "rank": 13,
      "name": "Claude Fable 5.1 (Adaptive Reasoning, Xhigh Effort, Default Fallback)",
      "creator": "Anthropic",
      "slug": "claude-fable-5-1-xhigh",
      "detailsUrl": "/models/claude-fable-5-1-xhigh",
      "value": 89.79120879120879
    },
    {
      "rank": 14,
      "name": "K2 Horizon 375B A23B",
      "creator": "MBZUAI Institute of Foundation Models",
      "slug": "k2-horizon-375b-a23b",
      "detailsUrl": "/models/k2-horizon-375b-a23b",
      "value": 84.1685393258427
    },
    {
      "rank": 15,
      "name": "Kimi K3 (max)",
      "creator": "Kimi",
      "slug": "kimi-k3",
      "detailsUrl": "/models/kimi-k3",
      "value": 82.63333333333334
    },
    {
      "rank": 16,
      "name": "Qwen3.8 27B (xhigh)",
      "creator": "Alibaba",
      "slug": "qwen3-8-27b",
      "detailsUrl": "/models/qwen3-8-27b",
      "value": 82.3
    },
    {
      "rank": 17,
      "name": "Inkling (xhigh)",
      "creator": "Thinking Machines",
      "slug": "inkling",
      "detailsUrl": "/models/inkling",
      "value": 81.46153846153847
    },
    {
      "rank": 18,
      "name": "Claude Opus 5 (Adaptive Reasoning, High Effort)",
      "creator": "Anthropic",
      "slug": "claude-opus-5-high",
      "detailsUrl": "/models/claude-opus-5-high",
      "value": 76.31868131868131
    },
    {
      "rank": 19,
      "name": "GPT-5.6 Luna (max)",
      "creator": "OpenAI",
      "slug": "gpt-5-6-luna",
      "detailsUrl": "/models/gpt-5-6-luna",
      "value": 72.74725274725274
    },
    {
      "rank": 20,
      "name": "Claude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback)",
      "creator": "Anthropic",
      "slug": "claude-fable-5",
      "detailsUrl": "/models/claude-fable-5",
      "value": 67.06593406593407
    }
  ]
}