OpenVLM MMMU - Literature: leaderboard
Metric: Accuracy (%). Source: huggingface.co. 284 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5 Nano | 93.3 |
| 2 | Grok 2 (1212) | 93.3 |
| 3 | Gemini 1.5 Flash | 93.3 |
| 4 | GPT-4o ChatGPT | 93.3 |
| 5 | Gemini 2.5 Pro | 90 |
| 6 | GPT-5 | 90 |
| 7 | GPT-4.1 Mini | 90 |
| 8 | InternVL3-8B | 90 |
| 9 | InternVL2-8B | 90 |
| 10 | Qwen 2 VL 72B | 90 |
| 11 | GPT-4.1 Nano | 90 |
| 12 | Gemini 1.5 Flash (002) | 90 |
| 13 | InternVL2.5-78B | 90 |
| 14 | Gemini 2.0 Flash | 90 |
| 15 | GPT-4.1 (2025-04-14) | 90 |
Interactive version: theaggregate.ai/benchmark?slug=openvlm-mmmu-literature · How It Works · Data refreshed daily, snapshot 2026-09-05.