Skip to content
Yardstick
  1. Benchmarks
  2. Artificial Analysis Intelligence Index

Artificial Analysis Intelligence Index

One composite score of general capability, built from ten evaluations spanning agents, coding, general knowledge and scientific reasoning.

Overall capabilityRun by Artificial AnalysisOfficial leaderboard (opens in a new tab)

Top score
57.6Claude Opus 5.5
Median of field
40.929 models scored
Data as of
Oct 8, 2026Artificial Analysis leaderboard

Standings

Bars share one axis from 0 to 100.

  1. Rank 1
    Claude Opus 5.557.6

    Anthropic · $4 / $20New

  2. Rank 2
    Claude Sonnet 5.556

    Anthropic · $2 / $10New

  3. Rank 3
    Claude Fable 5.153.4

    Anthropic · $10 / $50New

  4. Rank 4
    GPT-6 Astra52.7

    OpenAI · $10 / $50New

  5. Rank 5
    Gemini 4 Argon52.6

    Google · $2 / $10NewPreview

  6. Rank 6
    GPT-6.1 Sol51.8

    OpenAI · $2 / $10New

  7. Rank 7
    Muse Spark 1.348.1

    Meta · $1.25 / $4.25New

  8. Rank 8
    Grok 4.746.5

    SpaceXAI · $2 / $6New

  9. Rank 9
    MiMo-V2.6-Pro46.3

    Xiaomi · $0.435 / $0.87NewOpen weights

  10. Rank 10
    Qwen3.8-Max45.4

    Alibaba · $2 / $6

  11. Rank 11
    GLM-5.344.8

    Z.ai · $1.40 / $4.40NewOpen weights

  12. Rank 12
    Kimi K343.6

    Moonshot AI · $3 / $15Open weights

  13. Rank 13
    Claude Haiku 5.543.4

    Anthropic · $0.10 / $0.50New

  14. Rank 14
    GLM-5.3-Flash41.8

    Z.ai · $0.15 / $0.50NewOpen weights

  15. Rank 15
    Gemini 3.8 Flash40.9

    Google · $0.75 / $3.75New

  16. Rank 16
    Qwen3.8-Flash-Next39.8

    Alibaba · $0.15 / $0.47NewOpen weights

  17. Rank 17
    DeepSeek-V4.1-Flash39.5

    DeepSeek · $0.30 / $1.20NewOpen weights

  18. Rank 18
    Mistral Large 438.4

    Mistral AI · $1.36 / $4.18NewPreview

  19. Rank 19
    GPT-6 Luna38.1

    OpenAI · $0.10 / $0.50New

  20. Rank 20
    DeepSeek-V4-Pro36

    DeepSeek · $1.32 / $3.96NewOpen weights

  21. Rank 21
    Qwen3.8-27B33.7

    Alibaba · $0.50 / $3NewOpen weights

  22. Rank 22
    Gemini 3.1 Pro29.7

    Google · $2 / $12Preview

  23. Rank 23
    MiniMax M329.2

    MiniMax · $0.30 / $1.20Open weights

  24. Rank 24
    Kimi K2.7 Code25.8

    Moonshot AI · $0.95 / $4Open weights

  25. Rank 25
    Gemini 3.5 Flash-Lite22.2

    Google · $0.30 / $2.50

  26. Rank 26
    Muse Glimmer 30B17.5

    Meta · Self-hostedNewOpen weights

  27. Rank 27
    Gemma 4 31B14.7

    Google · Self-hostedOpen weights

  28. Rank 28
    Mistral Medium 3.514.2

    Mistral AI · $1.50 / $7.50Open weights

  29. Rank 29
    Mistral Small 411.3

    Mistral AI · $0.15 / $0.60Open weights

What it measures

General capability across four areas, weighted Agents 30%, Coding 20%, General 30% and Scientific reasoning 20%. Version 4.3 combines ten evaluations, among them GDPval-AA, Terminal-Bench 4.0, SciCode, Humanity's Last Exam, CritPt and Artificial Analysis's own Briefcase, Omniscience and long-context reasoning tests.

How it is scored

Artificial Analysis runs every model itself through one harness and reports a weighted average from 0 to 100. Each reasoning-effort setting is tested separately; this site shows a model's best setting and lists the rest on its page.

Read it with care

Version 4.3 (September 2026) swapped several component tests, so scores are not comparable with earlier versions of the index: the frontier now sits in the 50s. Artificial Analysis flags some values for older models as estimates.

Other benchmarks

  • Epoch Capabilities Index
  • LiveBench
  • Terminal-Bench 4.0
  • SWE-Bench Pro (V2 Hard)
  • Arena WebDev
  • GDPval-AA
  • Vending-Bench 2
  • ARC-AGI-2
  • ARC-AGI-3
  • FrontierMath (Tiers 1–3)
  • MathArena
  • SimpleBench
  • Humanity's Last Exam (Diamond)
  • GPQA Diamond
  • Arena Vision
  • Arena Text
Yardstick

A plain reading of how today’s SI models compare. Scores are taken from each benchmark’s own leaderboard wherever one exists; anything a lab reported about its own model is marked.

Data as of Oct 8, 2026.

Explore

  • Models
  • Compare
  • Benchmarks
  • Value
  • Labs
  • Methodology & sources

Sources

  • Artificial Analysis (opens in a new tab)
  • Arena (opens in a new tab)
  • Epoch AI Benchmarking Hub (opens in a new tab)
  • ARC Prize (opens in a new tab)
  • Terminal-Bench (opens in a new tab)
  • Scale AI Labs (opens in a new tab)

Yardstick is an independent reference and is not affiliated with, sponsored by or endorsed by any lab listed. Product names and logos are trademarks of their respective owners and are used only to identify them. Benchmarks are imperfect proxies: check a model on your own task before you commit to it.

  • Models
  • Compare
  • Benchmarks
  • Value
  • Labs
  • Models
  • Compare
  • Benchmarks
  • Value
  • Labs
  • Models
  • Compare
  • Benchmarks
  • Value
  • Labs