Skip to content
Model Index
  • Overview
  • Models
  • Compare
  • Benchmarks
  • Finder
LabsMethodology
  1. Benchmarks
  2. Vision

MMMU-Pro

Independent testArtificial Analysis · updated Oct 8, 2026

What it measures

1,730 college-level questions that can only be answered by reading an image — charts, diagrams, figures — with ten answer options.

How to read it

The standard read on visual reasoning. Text-only models have no score.

Open at Artificial Analysis
Best
87.7%
Claude Opus 5.5
Median
78.2%
Grok 4.7

Results

27 of 41 tracked models have a result.

Filter in explorer
  1. 1
    Claude Opus 5.5

    Anthropic · Max effort

    87.7%
  2. 2
    GPT-6 Astra

    OpenAI · Max effort

    86.9%
  3. 3
    GPT-6.1 Sol

    OpenAI · Max effort

    86.0%
  4. 4
    Gemini 3.8 Flash

    Google DeepMind · High effort

    85.6%
  5. 5
    Qwen3.8-Max

    Alibaba

    82.8%
  6. 6
    Gemini 3.1 Pro

    Google DeepMind

    82.4%
  7. 7
    Muse Spark 1.3

    Meta · xHigh effort

    82.0%
  8. 8
    GPT-5.6 Terra

    OpenAI · Max effort

    80.7%
  9. 9
    Kimi K3Open

    Moonshot AI · Max effort

    80.5%
  10. 10
    Qwen3.8-FlashOpen

    Alibaba

    79.8%
  11. 11
    GPT-6 Luna

    OpenAI · Max effort

    79.7%
  12. 12
    Gemini 3.5 Flash-Lite

    Google DeepMind

    79.0%
  13. 13
    MiniMax-M3Open

    MiniMax

    78.6%
  14. 14
    Grok 4.7

    SpaceXAI · Low effort

    78.2%
  15. 15
    DeepSeek V4.1 FlashOpen

    DeepSeek · Max effort

    77.0%
  16. 16
    Mistral Large 4

    Mistral AI

    76.4%
  17. 16
    Step 5 Preview

    StepFun

    76.4%
  18. 18
    Qwen3.8-27BOpen

    Alibaba · xHigh effort

    76.3%
  19. 19
    Muse Glimmer 30BOpen

    Meta · High effort

    74.3%
  20. 20
    TM
    Inkling-SmallOpen

    Thinking Machines Lab

    74.0%
  21. 21
    TM
    InklingOpen

    Thinking Machines Lab · xHigh effort

    73.5%
  22. 22
    Gemma 4 31BOpen

    Google DeepMind · Reasoning effort

    73.4%
  23. 23
    MiMo-V2.6-FlashOpen

    Xiaomi

    73.1%
  24. 24
    Mistral Medium 3.5Open

    Mistral AI

    64.9%
  25. 25
    Nova 2 Lite

    Amazon · High effort

    63.8%
  26. 26
    Command A+Open

    Cohere

    63.2%
  27. 27
    Mistral Small 4Open

    Mistral AI · Reasoning effort

    56.8%
14 tracked models without a result

Claude Fable 5.1, Claude Haiku 5.5, Claude Sonnet 5.5, DeepSeek V4 Pro, Gemini 4 Argon, GLM-5.3, GLM-5.3-Flash, gpt-oss-120b, Hy3, Hy4 preview, Kimi K2.7 Code, MAI-Thinking-1, MiMo-V2.6-Pro, Nemotron 3 Ultra.

Missing means the evaluator has not published a run, not that the model would score poorly.

Model Index

An independent snapshot of frontier SI models and how they perform on third-party benchmarks.

Data as of Oct 8, 2026

Explore

  • All models
  • Compare
  • Benchmarks
  • Model finder
  • Labs
  • Methodology & sources

Sources

Benchmark results come from Artificial Analysis, Arena, Vals AI and ARC Prize. Specifications and prices come from each lab's own documentation.

Not affiliated with any lab or evaluator. Names and logos are trademarks of their respective owners and are used only to identify them. Scores change often — check the linked source before making a decision that depends on one.

  • Overview
  • Models
  • Compare
  • Benchmarks
  • Finder
  • Overview
  • Models
  • Compare
  • Benchmarks
  • Finder
  • Overview
  • Models
  • Compare
  • Benchmarks
  • Finder