Skip to content
Yardstick
  1. Benchmarks
  2. LiveBench

LiveBench

A broad test whose questions are regularly replaced, so answers cannot simply have been memorised from training data.

Overall capabilityRun by LiveBenchOfficial leaderboard (opens in a new tab)

Top score
83.4%Claude Fable 5.1
Median of field
77%23 models scored
Data as of
Oct 8, 2026LiveBench data files, 2026-06-25 question set

Standings

Bars share one axis from 0% to 100%.

  1. Rank 1
    Claude Fable 5.183.4%

    Anthropic · $10 / $50New

  2. Rank 2
    Claude Opus 5.583.2%

    Anthropic · $4 / $20New

  3. Rank 3
    GPT-6 Astra82.2%

    OpenAI · $10 / $50New

  4. Rank 4
    GPT-6.1 Sol81.6%

    OpenAI · $2 / $10New

  5. Rank 5
    Muse Spark 1.381.6%

    Meta · $1.25 / $4.25New

  6. Rank 6
    DeepSeek-V4.1-Flash81.1%

    DeepSeek · $0.30 / $1.20NewOpen weights

  7. Rank 7
    Kimi K379.2%

    Moonshot AI · $3 / $15Open weights

  8. Rank 8
    Qwen3.8-Max78.5%

    Alibaba · $2 / $6

  9. Rank 9
    Claude Sonnet 5.577.8%

    Anthropic · $2 / $10New

  10. Rank 10
    DeepSeek-V4-Pro77.4%

    DeepSeek · $1.32 / $3.96NewOpen weights

  11. Rank 11
    Grok 4.777.4%

    SpaceXAI · $2 / $6New

  12. Rank 12
    Gemini 3.1 Pro77%

    Google · $2 / $12Preview

  13. Rank 13
    Qwen3.8-Flash-Next76.2%

    Alibaba · $0.15 / $0.47NewOpen weights

  14. Rank 14
    GLM-5.376.1%

    Z.ai · $1.40 / $4.40NewOpen weights

  15. Rank 15
    Gemini 3.8 Flash75.8%

    Google · $0.75 / $3.75New

  16. Rank 16
    Qwen3.8-27B75.3%

    Alibaba · $0.50 / $3NewOpen weights

  17. Rank 17
    Claude Haiku 5.572.1%

    Anthropic · $0.10 / $0.50New

  18. Rank 18
    GPT-6 Luna72%

    OpenAI · $0.10 / $0.50New

  19. Rank 19
    Mistral Large 471.8%

    Mistral AI · $1.36 / $4.18NewPreview

  20. Rank 20
    GLM-5.3-Flash71.6%

    Z.ai · $0.15 / $0.50NewOpen weights

  21. Rank 21
    Kimi K2.7 Code68.4%

    Moonshot AI · $0.95 / $4Open weights

  22. Rank 22
    MiniMax M367.3%

    MiniMax · $0.30 / $1.20Open weights

  23. Rank 23
    Gemini 3.5 Flash-Lite63.9%

    Google · $0.30 / $2.50

Not reported for LiveBench · 6

  • ·
    Gemini 4 Argon—Not reported

    Google · $2 / $10NewPreview

  • ·
    Gemma 4 31B—Not reported

    Google · Self-hostedOpen weights

  • ·
    MiMo-V2.6-Pro—Not reported

    Xiaomi · $0.435 / $0.87NewOpen weights

  • ·
    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

What it measures

Seven areas — reasoning, coding, agentic coding, mathematics, data analysis, language and instruction following — each made of several task types with objective, automatically checked answers.

How it is scored

The global score is the mean of the seven category averages. These numbers use the 25 June 2026 question set, computed from the site's own data files.

Read it with care

Models released after a question set's date are added to the same table, so the guarantee that questions are unseen weakens the longer a set stays live.

Other benchmarks

  • Artificial Analysis Intelligence Index
  • Epoch Capabilities Index
  • 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