Back to BenchmarksGoogle
🌐
MMMLU
Multilingual MMLU
Multilingual version of MMLU across multiple languages
🌐 Multilingual
Best: 91.8%
5 Models Evaluated
Champion
Gemini 3.1 Pro
91.8%
Best Score
Model Rankings
All models ranked by score on MMMLU. Scores are relative to the best score of 91.8%.
| # | Model | Provider | Score | vs Best |
|---|---|---|---|---|
| 1 | Gemini 3.1 Pro Elo 1492 | 91.8% | Best | |
| 2 | Claude Opus 4.6 Elo 1503 | Anthropic | 91.1% | -0.7% |
| 3 | Qwen 3.5 Elo 1450 | Alibaba | 88.5% | -3.3% |
| 4 | Llama 4 Maverick Elo 1328 | Meta | 84.6% | -7.2% |
| 5 | DeepSeek V4 Pro (Max) | DeepSeek | 83.5% | -8.3% |
Score Distribution
Gemini 3.1 Pro
91.8%
Claude Opus 4.6
91.1%
Qwen 3.5
88.5%
Llama 4 Maverick
84.6%
DeepSeek V4 Pro (Max)
83.5%
Run This Benchmark
Learn how to evaluate your own models on MMMLU.
MMMLU 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.