Skip to content
Yardstick
  1. Benchmarks
  2. Arena Text

Arena Text

Blind head-to-head votes from millions of real conversations: which answer did people prefer?

Human preferenceRun by Arena (formerly LMArena)Official leaderboard (opens in a new tab)

Top score
1,525Gemini 4 Argon
Median of field
1,47724 models scored
Data as of
Oct 2, 2026Arena leaderboard

Standings

Dots mark each rating on a shared scale from 1,400 to 1,540; ratings have no meaningful zero, so bar lengths would mislead.

  1. Rank 1
    Gemini 4 Argon1,525

    Google · $2 / $10NewPreview

  2. Rank 2
    Claude Opus 5.51,504

    Anthropic · $4 / $20New

  3. Rank 3
    Claude Fable 5.11,501

    Anthropic · $10 / $50New

  4. Rank 4
    Gemini 3.8 Flash1,495

    Google · $0.75 / $3.75New

  5. Rank 5
    Muse Spark 1.31,494

    Meta · $1.25 / $4.25New

  6. Rank 6
    Kimi K31,488

    Moonshot AI · $3 / $15Open weights

  7. Rank 7
    Gemini 3.1 Pro1,487

    Google · $2 / $12Preview

  8. Rank 8
    GPT-6.1 Sol1,483

    OpenAI · $2 / $10New

  9. Rank 9
    Qwen3.8-Max1,482

    Alibaba · $2 / $6

  10. Rank 10
    MiMo-V2.6-Pro1,480

    Xiaomi · $0.435 / $0.87NewOpen weights

  11. Rank 11
    GLM-5.31,478

    Z.ai · $1.40 / $4.40NewOpen weights

  12. Rank 12
    GPT-6 Astra1,477

    OpenAI · $10 / $50New

  13. Rank 13
    DeepSeek-V4.1-Flash1,474

    DeepSeek · $0.30 / $1.20NewOpen weights

  14. Rank 14
    GLM-5.3-Flash1,473

    Z.ai · $0.15 / $0.50NewOpen weights

  15. Rank 15
    Claude Sonnet 5.51,471

    Anthropic · $2 / $10New

  16. Rank 16
    DeepSeek-V4-Pro1,464

    DeepSeek · $1.32 / $3.96NewOpen weights

  17. Rank 17
    Gemini 3.5 Flash-Lite1,455

    Google · $0.30 / $2.50

  18. Rank 18
    Gemma 4 31B1,453

    Google · Self-hostedOpen weights

  19. Rank 19
    GPT-6 Luna1,443

    OpenAI · $0.10 / $0.50New

  20. Rank 20
    Grok 4.71,442

    SpaceXAI · $2 / $6New

  21. Rank 21
    MiniMax M31,440

    MiniMax · $0.30 / $1.20Open weights

  22. Rank 22
    Qwen3.8-27B1,438

    Alibaba · $0.50 / $3NewOpen weights

  23. Rank 23
    Mistral Medium 3.51,427

    Mistral AI · $1.50 / $7.50Open weights

  24. Rank 24
    Muse Glimmer 30B1,425

    Meta · Self-hostedNewOpen weights

Not reported for Arena Text · 5

  • ·
    Claude Haiku 5.5—Not reported

    Anthropic · $0.10 / $0.50New

  • ·
    Kimi K2.7 Code—Not reported

    Moonshot AI · $0.95 / $4Open weights

  • ·
    Mistral Large 4—Not reported

    Mistral AI · $1.36 / $4.18NewPreview

  • ·
    Mistral Small 4—Not reported

    Mistral AI · $0.15 / $0.60Open weights

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

    Alibaba · $0.15 / $0.47NewOpen weights

What it measures

Overall human preference in open-ended chat — writing, analysis, coding questions and everyday tasks.

How it is scored

Two anonymous models answer the same prompt and the user picks a winner. More than eight million votes are fitted to an Elo-style rating, with Arena's style control applied.

Read it with care

It measures what people like, which rewards tone and formatting as well as accuracy. Adjacent ranks usually overlap within their confidence intervals, and preliminary scores rest on few votes.

Other benchmarks

  • Artificial Analysis Intelligence Index
  • 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
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