OpenVLM MMMU - Health and Medicine: leaderboard
OpenCompass OpenVLM MMMU validation subcategory score for visual college-level questions in health and medicine.
Metric: Accuracy (%). Source: huggingface.co. Saturation forecast: Around January 2027. 284 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5 | 79.3 | #98 |
| 2 | GPT-5 Mini (2025-08-07) | 79.3 | #172 |
| 3 | GPT-4.1 (2025-04-14) | 78.7 | #250 |
| 4 | GPT-4.5 | 75.3 | #274 |
| 5 | GPT-4o ChatGPT | 75.3 | |
| 6 | Gemini 2.5 Pro | 74.7 | #152 |
| 7 | GPT-5 Nano | 74.7 | #425 |
| 8 | InternVL3-78B | 74 | #352 |
| 9 | Claude 3.7 Sonnet | 74 | #249 |
| 10 | Claude 3.5 Sonnet | 74 | #350 |
| 11 | InternVL2.5-78B | 74 | #431 |
| 12 | InternVL3-38B | 71.3 | #405 |
| 13 | Claude 3.5 Sonnet (20241022) | 70.7 | #291 |
| 14 | Gemini 1.5 Pro (002) | 70.7 | #349 |
| 15 | Grok 2 (1212) | 70.7 | #496 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=openvlm-mmmu-health-and-medicine · How It Works · Data refreshed daily, snapshot 2026-10-20.