Back to Benchmarks
👁️
MMMU-Pro
MMMU-Professional
Robust multimodal reasoning across 30 disciplines, vision-only input
👁️ Multimodal
Best: 81.2%
7 Models Evaluated
Champion
GPT-5.5
OpenAI
81.2%
Best Score
Model Rankings
All models ranked by score on MMMU-Pro. Scores are relative to the best score of 81.2%.
| # | Model | Provider | Score | vs Best |
|---|---|---|---|---|
| 1 | GPT-5.5 Elo 1478 | OpenAI | 81.2% | Best |
| 2 | GPT-5.4 Elo 1463 | OpenAI | 81.2% | +0.0% |
| 3 | GPT-5.4 Pro Elo 1480 | OpenAI | 81.2% | +0.0% |
| 4 | Gemini 3.1 Pro Elo 1492 | 81% | -0.2% | |
| 5 | Qwen 3.5 Elo 1450 | Alibaba | 79% | -2.2% |
| 6 | Claude Opus 4.6 Elo 1503 | Anthropic | 77.3% | -3.9% |
| 7 | Claude Sonnet 4.6 Elo 1460 | Anthropic | 75.6% | -5.6% |
Score Distribution
GPT-5.5
81.2%
GPT-5.4
81.2%
GPT-5.4 Pro
81.2%
Gemini 3.1 Pro
81%
Qwen 3.5
79%
Claude Opus 4.6
77.3%
Claude Sonnet 4.6
75.6%
Run This Benchmark
Learn how to evaluate your own models on MMMU-Pro.
MMMU-Pro is a standardized benchmark for evaluating AI model capabilities.
To run this benchmark on your own models, visit the official benchmark page or leaderboard for detailed instructions, code, and evaluation scripts.
Data compiled from official model cards, public leaderboards, and independent evaluations.