{
  "url": "https://www.aibattle.space/rankings/eval-gdpval-gdpval-aa-v2-average-turns-per-task",
  "dataset": "GDPval-AA v2: Average Turns per Task",
  "family": "tab::eval-gdpval",
  "familyLabel": "eval gdpval",
  "sourceRowCount": 20,
  "window": 20,
  "metricKeys": [
    "averageTurnsPerTask"
  ],
  "rebuilt": "2026-09-19",
  "stats": {
    "median": 56.074958298582146,
    "mean": 56.4470669706624,
    "min": 37.0090909090909,
    "max": 80.9305555555556,
    "spread": 43.9214646464647,
    "gapFirstToSecond": 2.693112633181201,
    "gapFirstToSecondPercent": 3.44,
    "creators": 12
  },
  "entries": [
    {
      "rank": 1,
      "name": "GLM-5.3-Flash",
      "creator": "Z AI",
      "slug": "glm-5-3-flash",
      "detailsUrl": "/models/glm-5-3-flash",
      "value": 80.9305555555556
    },
    {
      "rank": 2,
      "name": "Gemini 3.8 Flash (high)",
      "creator": "Google",
      "slug": "gemini-3-8-flash",
      "detailsUrl": "/models/gemini-3-8-flash",
      "value": 78.2374429223744
    },
    {
      "rank": 3,
      "name": "Muse Spark 1.3 (max)",
      "creator": "Meta",
      "slug": "muse-spark-1-3",
      "detailsUrl": "/models/muse-spark-1-3",
      "value": 74.4861111111111
    },
    {
      "rank": 4,
      "name": "DeepSeek V4.1 Flash (Reasoning, Max Effort)",
      "creator": "DeepSeek",
      "slug": "deepseek-v4-1-flash",
      "detailsUrl": "/models/deepseek-v4-1-flash",
      "value": 72.1324200913242
    },
    {
      "rank": 5,
      "name": "GLM-5.3 (max)",
      "creator": "Z AI",
      "slug": "glm-5-3",
      "detailsUrl": "/models/glm-5-3",
      "value": 69.9862385321101
    },
    {
      "rank": 6,
      "name": "Qwen3.8 2.4T A95B",
      "creator": "Alibaba",
      "slug": "qwen3-8-2-4t-a95b",
      "detailsUrl": "/models/qwen3-8-2-4t-a95b",
      "value": 66.963133640553
    },
    {
      "rank": 7,
      "name": "Muse Spark 1.3 (xhigh)",
      "creator": "Meta",
      "slug": "muse-spark-1-3-xhigh",
      "detailsUrl": "/models/muse-spark-1-3-xhigh",
      "value": 63.15
    },
    {
      "rank": 8,
      "name": "Claude Opus 5 (Adaptive Reasoning, Max Effort)",
      "creator": "Anthropic",
      "slug": "claude-opus-5",
      "detailsUrl": "/models/claude-opus-5",
      "value": 60.4703196347032
    },
    {
      "rank": 9,
      "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": 60.3926940639269
    },
    {
      "rank": 10,
      "name": "Qwen3.8 27B (xhigh)",
      "creator": "Alibaba",
      "slug": "qwen3-8-27b",
      "detailsUrl": "/models/qwen3-8-27b",
      "value": 57.9908256880734
    },
    {
      "rank": 11,
      "name": "MiniMax-M3",
      "creator": "MiniMax",
      "slug": "minimax-m3",
      "detailsUrl": "/models/minimax-m3",
      "value": 54.1590909090909
    },
    {
      "rank": 12,
      "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": 50.7397260273973
    },
    {
      "rank": 13,
      "name": "Claude Opus 5 (Adaptive Reasoning, Xhigh Effort)",
      "creator": "Anthropic",
      "slug": "claude-opus-5-xhigh",
      "detailsUrl": "/models/claude-opus-5-xhigh",
      "value": 50.6727272727273
    },
    {
      "rank": 14,
      "name": "GPT-5.6 Sol (max)",
      "creator": "OpenAI",
      "slug": "gpt-5-6-sol",
      "detailsUrl": "/models/gpt-5-6-sol",
      "value": 45.2054794520548
    },
    {
      "rank": 15,
      "name": "K2 Horizon 375B A23B",
      "creator": "MBZUAI Institute of Foundation Models",
      "slug": "k2-horizon-375b-a23b",
      "detailsUrl": "/models/k2-horizon-375b-a23b",
      "value": 44.9909090909091
    },
    {
      "rank": 16,
      "name": "Inkling (xhigh)",
      "creator": "Thinking Machines",
      "slug": "inkling",
      "detailsUrl": "/models/inkling",
      "value": 43.1045454545455
    },
    {
      "rank": 17,
      "name": "Grok 4.6 (xhigh)",
      "creator": "SpaceXAI",
      "slug": "grok-4-6-xhigh",
      "detailsUrl": "/models/grok-4-6-xhigh",
      "value": 41.9136363636364
    },
    {
      "rank": 18,
      "name": "DeepSeek V4 Pro 0813 (Reasoning, Max Effort)",
      "creator": "DeepSeek",
      "slug": "deepseek-v4-pro",
      "detailsUrl": "/models/deepseek-v4-pro",
      "value": 38.8949771689498
    },
    {
      "rank": 19,
      "name": "Kimi K3 (max)",
      "creator": "Kimi",
      "slug": "kimi-k3",
      "detailsUrl": "/models/kimi-k3",
      "value": 37.5114155251142
    },
    {
      "rank": 20,
      "name": "Grok 4.6 (high)",
      "creator": "SpaceXAI",
      "slug": "grok-4-6",
      "detailsUrl": "/models/grok-4-6",
      "value": 37.0090909090909
    }
  ]
}