ARTIFICIAL ANALYSIS / INTELLIGENCE INDEX

AI大模型能力排行榜-2026年10月13日更新

每小时自动更新数据来源 Artificial Analysis ↗
排名 / 模型能力指数 ↓
  1. 01
    Claude Opus 5.5 (Max, Default Fallback)
    Anthropic
    57.6
  2. 02
    Claude Sonnet 5.5 (Max, Default Fallback)
    Anthropic
    56.0
  3. 03
    Muse Wonder Full (Max)
    Meta
    55.8
  4. 04
    Claude Fable 5.1 (Max, Default Fallback)
    Anthropic
    53.4
  5. 05
    GPT-6 Astra (Max)
    OpenAI
    52.7
  6. 06
    Gemini 4 Argon (High)
    Google
    52.6
  7. 07
    Muse Wonder (Max)
    Meta
    52.4
  8. 08
    GPT-6.1 Sol (Max)
    OpenAI
    51.8
  9. 09
    Claude Opus 5 (High)
    Anthropic · 历史评测
    48.1
  10. 10
    Muse Spark 1.3 (Max)
    Meta
    48.1
  11. 11
    Grok 4.7 (Xhigh)
    SpaceXAI
    46.4
  12. 12
    MiMo-V2.6-Pro
    Xiaomi
    46.3
  13. 13
    Qwen3.8 Max (0902)
    Alibaba
    45.4
  14. 14
    GLM-5.3 (Max)
    Z AI
    44.8
  15. 15
    Step 5 Preview
    StepFun
    43.7
  16. 16
    Kimi K3 (Max)
    Kimi
    43.6
  17. 17
    Claude Haiku 5.5 (Max)
    Anthropic
    43.4
  18. 18
    GPT-5.6 Sol (High)
    OpenAI · 历史评测
    42.3
  19. 19
    GLM-5.3-Flash
    Z AI
    41.8
  20. 20
    Gemini 3.8 Flash (High)
    Google
    40.9
  21. 21
    DeepSeek V4.1 Flash (Max)
    DeepSeek
    39.5
  22. 22
    Mistral Large 4 Preview
    Mistral
    38.4
  23. 23
    GPT-6 Luna (Max)
    OpenAI
    38.1
  24. 24
    GPT-5.6 Terra (High)
    OpenAI
    34.2
  25. 25
    Qwen3.8 27B (Xhigh)
    Alibaba
    33.7
  26. 26
    K2 Horizon 375B A23B
    Institute of Foundation Models
    30.5
  27. 27
    MiniMax-M3
    MiniMax
    29.2
  28. 28
    Inkling (Xhigh)
    Thinking Machines
    25.0
  29. 29
    Nemotron 3 Ultra 550B A55B (Reasoning)
    NVIDIA
    22.9
  30. 30
    Muse Glimmer (High)
    Meta
    17.5

统计范围：Artificial Analysis 默认展示模型及指定 High 配置，共 30 个模型配置。指数越高，综合能力越强；同分按原始精度排序。

每小时同步一次。点击模型名称查看原站评测。