Agents Directory
SkillsRankingsAgents
CategoriesModelsBenchmarksCompareAgent LeaderboardSkillsRankingsAgentsAbout
/Benchmarks
/Artificial Analysis Intelligence Index
Artificial Analysis Intelligence Index logo

Artificial Analysis Intelligence Index

Intelligence

The most-cited composite intelligence score: a 0–100 index combining knowledge, reasoning, math, coding, and agentic evaluations (GPQA Diamond, HLE, IFBench, SciCode, Terminal-Bench Hard, τ²-Bench, and more). Higher is better.

Official source
Leaderboard
#ModelScorePrice
  • 1
    ClaudeClaude Fable 5Adaptive Reasoning, Max Effort, Opus 4.8 Fallback
    59.9$20.00
  • 2
    ClaudeClaude Opus 4.8Adaptive Reasoning, Max Effort
    55.7$10.00
  • 3
    OpenAIGPT-5.5xhigh
    54.8$11.25
  • 4
    ClaudeClaude Opus 4.7Adaptive Reasoning, Max Effort
    53.5$10.00
  • 5
    OpenAIGPT-5.5high
    53.1$11.25
  • 6
    OpenAIGPT-5.4xhigh
    51.4$5.63
  • 7
    GLM-5.2 logoGLM-5.2max
    51.1$2.15
  • 8
    OpenAIGPT-5.5medium
    50.4$11.25
  • 9
    GeminiGemini 3.5 Flashhigh
    50.2$3.38
  • 10
    ClaudeClaude Sonnet 4.6Adaptive Reasoning, Max Effort
    47.2$6.00
  • 11
    GeminiGemini 3.1 Pro Preview
    46.5$4.50
  • 12
    QwenQwen3.7 Max
    46$3.75
  • 13
    GeminiGemini 3.5 Flashmedium
    45.4$3.38
  • 14
    MinimaxMiniMax-M3
    44.4$0.53
  • 15
    DeepSeekDeepSeek V4Reasoning, Max Effort
    44.3$0.54
  • 16
    OpenAIGPT-5.5low
    43.5$11.25
  • 17
    MoonshotAIKimi K2.6
    42.8$1.71
  • 18
    ClaudeClaude Opus 4.7Non-reasoning, High Effort
    42.7$10.00
  • 19
    OpenAIGPT-5.2xhigh
    42.2$4.81
  • 20
    MoonshotAIKimi K2.7 Code
    41.9$1.71
  • 21
    DeepSeekDeepSeek V4Reasoning, High Effort
    40.8$0.54
  • 22
    OpenAIGPT-5.2 Codexxhigh
    40.1$4.81
  • 23
    OpenAIGPT-5.4 Minixhigh
    40$1.69
  • 24
    GeminiGemini 3 Prohigh
    39.6$4.50
  • 25
    OpenAIGPT-5.4low
    39.1$5.91
  • 26
    MoonshotAIKimi K2.5Reasoning
    38.1$1.19
  • 27
    OpenAIGPT-5.2medium
    38$4.81
  • 28
    GeminiGemini 3 FlashReasoning
    37.8$1.13
  • 29
    GrokGrok 4.3high
    37.6$1.56
  • 30
    GrokGrok 4.3medium
    36$1.56
  • 31
    ClaudeClaude Sonnet 4.6Non-reasoning, High Effort
    35.9$6.00
  • 32
    GrokGrok 4.3low
    35.4$1.56
  • 33
    OpenAIGPT-5.5Non-reasoning
    35.4$11.25
  • 34
    GeminiGemini 3.5 Flashminimal
    34.9$3.38
  • 35
    MoonshotAIKimi K2.6Non-reasoning
    34.6$1.71
  • 36
    ClaudeClaude Sonnet 4.6Non-reasoning, Low Effort
    34.3$6.00
  • 37
    OpenAIGPT-5.5May 2026
    33.5$11.25
  • 38
    GeminiGemini 3 Prolow
    33.1$4.50
  • 39
    DeepSeekDeepSeek V4Non-reasoning
    31.2$0.54
  • 40
    OpenAIGPT-5.4 Minimedium
    29.8$1.69
  • 41
    ClaudeClaude Haiku 4.5Reasoning
    29.6$2.00
  • 42
    MoonshotAIKimi K2.5Non-reasoning
    29.4$1.20
  • 43
    OpenAIGPT-5.4Non-reasoning
    27.7$5.91
  • 44
    GeminiGemini 3 FlashNon-reasoning
    27.4$1.13
  • 45
    OpenAIGPT-5.2Non-reasoning
    26$4.81
  • 46
    GrokGrok 4.3Non-reasoning
    24.8$1.56
  • 47
    ClaudeClaude Haiku 4.5Non-reasoning
    23.7$2.00
  • 48
    OpenAIGPT-5.4 MiniNon-Reasoning
    16.6$1.69
  • 49
    MistralMistral Large 3
    15.9$0.75
Sources:
AA Data API (models endpoint)Artificial Analysis Intelligence Index
Used in rankings:
Best cheap AI modelsBest Chinese AI modelsBest Qwen models
Share:
Details:

  • Category

    Intelligence

  • Artificial Analysis logoCreated by

    Artificial Analysis

  • Models tested

    23

  • Configs tested

    49

  • Leader

    ClaudeClaude Fable 5

  • Top score

    59.9

Updated June 2026

Browse:SkillsRankingsModelsBenchmarksProvidersAgentsAgent LeaderboardCompareCategories
Quick Links:AboutBlog

© 2026 Agents Directory