CMMMU — leaderboard
Metric: Test Overall (%). Source: cmmmu-benchmark.github.io. 35 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-4o | 53.1 |
| 2 | GPT-4 | 33.8 |
| 3 | flan-t5-xxl | 31.2 |
| 4 | vicuna-13B | 31 |
| 5 | Llama 2 7B | 28.7 |
| 6 | Qwen-7B | 25.1 |
| 7 | Baichuan-7B | 24.3 |
| 8 | Yi 6B (Base) | 24.2 |
Interactive version: theaggregate.ai/benchmark?slug=cmmmu · How the rankings work · Data refreshed daily, snapshot 2026-07-22.