Skip to content
Yardstick
  1. Benchmarks
  2. Arena Vision

Arena Vision

Blind head-to-head votes on answers to prompts that include an image.

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

Top score
1,301Qwen3.8-Max
Median of field
1,27815 models scored
Data as of
Oct 2, 2026Arena leaderboard

Standings

Dots mark each rating on a shared scale from 1,180 to 1,320; ratings have no meaningful zero, so bar lengths would mislead.

  1. Rank 1
    Qwen3.8-Max1,301

    Alibaba · $2 / $6

  2. Rank 2
    GPT-6.1 Sol1,291

    OpenAI · $2 / $10New

  3. Rank 3
    Gemini 3.8 Flash1,290

    Google · $0.75 / $3.75New

  4. Rank 3
    Muse Spark 1.31,290

    Meta · $1.25 / $4.25New

  5. Rank 5
    Claude Fable 5.11,288

    Anthropic · $10 / $50New

  6. Rank 6
    GPT-6 Astra1,284

    OpenAI · $10 / $50New

  7. Rank 7
    Gemini 3.1 Pro1,279

    Google · $2 / $12Preview

  8. Rank 8
    GLM-5.3-Flash1,278

    Z.ai · $0.15 / $0.50NewOpen weights

  9. Rank 9
    Claude Sonnet 5.51,268

    Anthropic · $2 / $10New

  10. Rank 10
    Gemini 3.5 Flash-Lite1,264

    Google · $0.30 / $2.50

  11. Rank 11
    Gemma 4 31B1,262

    Google · Self-hostedOpen weights

  12. Rank 12
    MiMo-V2.6-Pro1,245

    Xiaomi · $0.435 / $0.87NewOpen weights

  13. Rank 13
    Qwen3.8-27B1,243

    Alibaba · $0.50 / $3NewOpen weights

  14. Rank 14
    MiniMax M31,237

    MiniMax · $0.30 / $1.20Open weights

  15. Rank 15
    Mistral Medium 3.51,199

    Mistral AI · $1.50 / $7.50Open weights

Not reported for Arena Vision · 14

  • ·
    Claude Haiku 5.5—Not reported

    Anthropic · $0.10 / $0.50New

  • ·
    Claude Opus 5.5—Not reported

    Anthropic · $4 / $20New

  • ·
    DeepSeek-V4-Pro—Not reported

    DeepSeek · $1.32 / $3.96NewOpen weights

  • ·
    DeepSeek-V4.1-Flash—Not reported

    DeepSeek · $0.30 / $1.20NewOpen weights

  • ·
    Gemini 4 Argon—Not reported

    Google · $2 / $10NewPreview

  • ·
    GLM-5.3—Not reported

    Z.ai · $1.40 / $4.40NewOpen weights

  • ·
    GPT-6 Luna—Not reported

    OpenAI · $0.10 / $0.50New

  • ·
    Grok 4.7—Not reported

    SpaceXAI · $2 / $6New

  • ·
    Kimi K2.7 Code—Not reported

    Moonshot AI · $0.95 / $4Open weights

  • ·
    Kimi K3—Not reported

    Moonshot AI · $3 / $15Open weights

  • ·
    Mistral Large 4—Not reported

    Mistral AI · $1.36 / $4.18NewPreview

  • ·
    Mistral Small 4—Not reported

    Mistral AI · $0.15 / $0.60Open weights

  • ·
    Muse Glimmer 30B—Not reported

    Meta · Self-hostedNewOpen weights

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

    Alibaba · $0.15 / $0.47NewOpen weights

What it measures

Which model people prefer when the question comes with a picture: reading charts and documents, describing scenes, reasoning about what is shown.

How it is scored

Two anonymous models answer the same image prompt and the user votes; votes are fitted to an Elo-style rating.

Read it with care

A preference measure with wide intervals for newer models. It says nothing about audio or video input.

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 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