{
  "url": "https://www.aibattle.space/rankings/eval-terminal-terminal-bench-v4-0-score",
  "dataset": "Terminal-Bench v4.0: Score",
  "family": "tab::eval-terminal",
  "familyLabel": "eval terminal",
  "sourceRowCount": 20,
  "window": 20,
  "metricKeys": [
    "Terminal-Bench v4.0"
  ],
  "rebuilt": "2026-09-19",
  "stats": {
    "median": 0.409090909090909,
    "mean": 0.3810606060606059,
    "min": 0.116161616161616,
    "max": 0.595959595959596,
    "spread": 0.47979797979798,
    "gapFirstToSecond": 0.005050505050504972,
    "gapFirstToSecondPercent": 0.85,
    "creators": 8
  },
  "entries": [
    {
      "rank": 1,
      "name": "GPT-6 Astra (xhigh)",
      "creator": "OpenAI",
      "slug": "gpt-6-astra-xhigh",
      "detailsUrl": "/models/gpt-6-astra-xhigh",
      "value": 0.595959595959596
    },
    {
      "rank": 2,
      "name": "GPT-6 Astra (max)",
      "creator": "OpenAI",
      "slug": "gpt-6-astra",
      "detailsUrl": "/models/gpt-6-astra",
      "value": 0.590909090909091
    },
    {
      "rank": 3,
      "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": 0.55050505050505
    },
    {
      "rank": 4,
      "name": "GPT-6 Astra (high)",
      "creator": "OpenAI",
      "slug": "gpt-6-astra-high",
      "detailsUrl": "/models/gpt-6-astra-high",
      "value": 0.54040404040404
    },
    {
      "rank": 5,
      "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": 0.52020202020202
    },
    {
      "rank": 6,
      "name": "Claude Fable 5.1 (Adaptive Reasoning, High Effort, Default Fallback)",
      "creator": "Anthropic",
      "slug": "claude-fable-5-1-high",
      "detailsUrl": "/models/claude-fable-5-1-high",
      "value": 0.52020202020202
    },
    {
      "rank": 7,
      "name": "GPT-6 Astra (medium)",
      "creator": "OpenAI",
      "slug": "gpt-6-astra-medium",
      "detailsUrl": "/models/gpt-6-astra-medium",
      "value": 0.494949494949495
    },
    {
      "rank": 8,
      "name": "Claude Opus 5 (Adaptive Reasoning, Max Effort)",
      "creator": "Anthropic",
      "slug": "claude-opus-5",
      "detailsUrl": "/models/claude-opus-5",
      "value": 0.48989898989899
    },
    {
      "rank": 9,
      "name": "Claude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback)",
      "creator": "Anthropic",
      "slug": "claude-fable-5",
      "detailsUrl": "/models/claude-fable-5",
      "value": 0.424242424242424
    },
    {
      "rank": 10,
      "name": "GLM-5.3 (max)",
      "creator": "Z AI",
      "slug": "glm-5-3",
      "detailsUrl": "/models/glm-5-3",
      "value": 0.419191919191919
    },
    {
      "rank": 11,
      "name": "GPT-5.6 Sol (max)",
      "creator": "OpenAI",
      "slug": "gpt-5-6-sol",
      "detailsUrl": "/models/gpt-5-6-sol",
      "value": 0.398989898989899
    },
    {
      "rank": 12,
      "name": "GPT-5.6 Terra (max)",
      "creator": "OpenAI",
      "slug": "gpt-5-6-terra",
      "detailsUrl": "/models/gpt-5-6-terra",
      "value": 0.353535353535354
    },
    {
      "rank": 13,
      "name": "Muse Spark 1.3 (max)",
      "creator": "Meta",
      "slug": "muse-spark-1-3",
      "detailsUrl": "/models/muse-spark-1-3",
      "value": 0.333333333333333
    },
    {
      "rank": 14,
      "name": "GLM-5.3-Flash",
      "creator": "Z AI",
      "slug": "glm-5-3-flash",
      "detailsUrl": "/models/glm-5-3-flash",
      "value": 0.328282828282828
    },
    {
      "rank": 15,
      "name": "DeepSeek V4.1 Flash (Reasoning, Max Effort)",
      "creator": "DeepSeek",
      "slug": "deepseek-v4-1-flash",
      "detailsUrl": "/models/deepseek-v4-1-flash",
      "value": 0.267676767676768
    },
    {
      "rank": 16,
      "name": "Grok 4.6 (high)",
      "creator": "SpaceXAI",
      "slug": "grok-4-6",
      "detailsUrl": "/models/grok-4-6",
      "value": 0.212121212121212
    },
    {
      "rank": 17,
      "name": "Gemini 3.8 Flash (high)",
      "creator": "Google",
      "slug": "gemini-3-8-flash",
      "detailsUrl": "/models/gemini-3-8-flash",
      "value": 0.196969696969697
    },
    {
      "rank": 18,
      "name": "DeepSeek V4 Pro 0813 (Reasoning, Max Effort)",
      "creator": "DeepSeek",
      "slug": "deepseek-v4-pro",
      "detailsUrl": "/models/deepseek-v4-pro",
      "value": 0.141414141414141
    },
    {
      "rank": 19,
      "name": "Kimi K3 (max)",
      "creator": "Kimi",
      "slug": "kimi-k3",
      "detailsUrl": "/models/kimi-k3",
      "value": 0.126262626262626
    },
    {
      "rank": 20,
      "name": "GPT-5.6 Luna (max)",
      "creator": "OpenAI",
      "slug": "gpt-5-6-luna",
      "detailsUrl": "/models/gpt-5-6-luna",
      "value": 0.116161616161616
    }
  ]
}