MMMU
Multimodal college-level reasoning.
Leaderboard
| # | Model | Provider | % | Evaluated | Source |
|---|---|---|---|---|---|
| 1 | Claude Opus 4.7 | Anthropic | 76.1% | — | |
| 2 | Gemini 2.0 Flash | Google DeepMind | 71.7% | — | |
| 3 | Claude 3.5 Sonnet | Anthropic | 70.4% | — | |
| 4 | GPT-4o | OpenAI | 69.1% | — |
Multimodal college-level reasoning.
| # | Model | Provider | % | Evaluated | Source |
|---|---|---|---|---|---|
| 1 | Claude Opus 4.7 | Anthropic | 76.1% | — | |
| 2 | Gemini 2.0 Flash | Google DeepMind | 71.7% | — | |
| 3 | Claude 3.5 Sonnet | Anthropic | 70.4% | — | |
| 4 | GPT-4o | OpenAI | 69.1% | — |