Skip to content
Yardstick
  1. Benchmarks
  2. Epoch Capabilities Index

Epoch Capabilities Index

A statistical composite that folds a model's results on dozens of benchmarks into one general-capability number.

Overall capabilityRun by Epoch AIOfficial leaderboard (opens in a new tab)

Top score
167.3Claude Opus 5.5
Median of field
156.716 models scored
Data as of
Sep 29, 2026Epoch AI Benchmarking Hub data export

Standings

Dots mark each rating on a shared scale from 144 to 170; ratings have no meaningful zero, so bar lengths would mislead.

  1. Rank 1
    Claude Opus 5.5167.3

    Anthropic · $4 / $20New

  2. Rank 2
    GPT-6 Astra166.4

    OpenAI · $10 / $50New

  3. Rank 3
    GPT-6.1 Sol166.1

    OpenAI · $2 / $10New

  4. Rank 4
    Claude Sonnet 5.5165

    Anthropic · $2 / $10New

  5. Rank 5
    Claude Fable 5.1164.7

    Anthropic · $10 / $50New

  6. Rank 6
    Kimi K3157.4

    Moonshot AI · $3 / $15Open weights

  7. Rank 7
    Muse Spark 1.3156.8

    Meta · $1.25 / $4.25New

  8. Rank 8
    Gemini 3.8 Flash156.7

    Google · $0.75 / $3.75New

  9. Rank 9
    Qwen3.8-Max156.4

    Alibaba · $2 / $6

  10. Rank 10
    GPT-6 Luna156.3

    OpenAI · $0.10 / $0.50New

  11. Rank 11
    GLM-5.3155.6

    Z.ai · $1.40 / $4.40NewOpen weights

  12. Rank 12
    DeepSeek-V4-Pro155.3

    DeepSeek · $1.32 / $3.96NewOpen weights

  13. Rank 13
    DeepSeek-V4.1-Flash154.9

    DeepSeek · $0.30 / $1.20NewOpen weights

  14. Rank 14
    Grok 4.7153.5

    SpaceXAI · $2 / $6New

  15. Rank 15
    GLM-5.3-Flash151.9

    Z.ai · $0.15 / $0.50NewOpen weights

  16. Rank 16
    MiniMax M3146.9

    MiniMax · $0.30 / $1.20Open weights

Not reported for Epoch Capabilities Index · 13

  • ·
    Claude Haiku 5.5—Not reported

    Anthropic · $0.10 / $0.50New

  • ·
    Gemini 3.1 Pro—Not reported

    Google · $2 / $12Preview

  • ·
    Gemini 3.5 Flash-Lite—Not reported

    Google · $0.30 / $2.50

  • ·
    Gemini 4 Argon—Not reported

    Google · $2 / $10NewPreview

  • ·
    Gemma 4 31B—Not reported

    Google · Self-hostedOpen weights

  • ·
    Kimi K2.7 Code—Not reported

    Moonshot AI · $0.95 / $4Open weights

  • ·
    MiMo-V2.6-Pro—Not reported

    Xiaomi · $0.435 / $0.87NewOpen weights

  • ·
    Mistral Large 4—Not reported

    Mistral AI · $1.36 / $4.18NewPreview

  • ·
    Mistral Medium 3.5—Not reported

    Mistral AI · $1.50 / $7.50Open weights

  • ·
    Mistral Small 4—Not reported

    Mistral AI · $0.15 / $0.60Open weights

  • ·
    Muse Glimmer 30B—Not reported

    Meta · Self-hostedNewOpen weights

  • ·
    Qwen3.8-27B—Not reported

    Alibaba · $0.50 / $3NewOpen weights

  • ·
    Qwen3.8-Flash-Next—Not reported

    Alibaba · $0.15 / $0.47NewOpen weights

What it measures

General capability, estimated jointly from many benchmarks so that models can be compared across a span of years in which any single benchmark saturates.

How it is scored

Epoch AI fits a statistical model across benchmark results and places every model on one linear scale. Only differences are meaningful, not the absolute value, which is why it is plotted here as a position rather than a bar.

Read it with care

The index inherits the weaknesses of its inputs, several of which are now at or near their ceiling. Intervals are a few points wide, so neighbours on the list are often indistinguishable.

Other benchmarks

  • Artificial Analysis Intelligence 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