Medical MM Leaderboard - Average: leaderboard
Medical Multimodal Leaderboard overall — average performance across medical visual-question-answering and multimodal medical reasoning datasets; accuracy (%).
Metric: Average Accuracy (%). Source: huggingface.co. Saturation forecast: Estimated already saturated. 19 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 2.5 Flash | 65.1 | #245 |
| 2 | GPT-4.1 | 63.4 | #248 |
| 3 | Claude Sonnet 4 | 61.5 | #199 |
| 4 | InternVL3-38B | 59.4 | #405 |
| 5 | InternVL3-14B | 58 | #505 |
| 6 | InternVL3-8B | 57.3 | #620 |
| 7 | MedGemma-4B-IT | 54.8 | #853 |
| 8 | Qwen 2.5 VL 7B | 50 | #643 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=medical-mm-leaderboard-average · How It Works · Data refreshed daily, snapshot 2026-10-20.