GMAI-MMBench: leaderboard
Each vision-language model answers zero-shot questions about medical images (38 imaging modalities, from public datasets and hospitals) on the held-out Test split; the score is the percentage of the split's single-choice questions answered correctly, as the authors report it. The Test split holds 21,281 questions; its multiple-choice items are scored separately and are not in this number. Unlike text-only medical exams such as MedQA, every question needs the image, and many point at a box, mask or contour in it.
Metric: Accuracy on single-choice questions (%). Source: github.com. Saturation forecast: Around February 2027. 24 models tracked. Updated 17 Nov 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | MedDr-40B | 43.69 | |
| 2 | DeepSeek-VL-7B-Chat | 43.43 | #3480 |
| 3 | MiniCPM-V-2 | 42.54 | #7982 |
| 4 | Qwen-VL-Max | 42.16 | #1332 |
| 5 | LLaVA-internLM2-7B | 40.45 | |
| 6 | InternVL-Chat-V1.5 | 39.73 | #3908 |
| 7 | OmniLMM-12B | 39.3 | #8047 |
| 8 | InternLM-XComposer2-VL-7B | 39.2 | |
| 9 | TransCore-M | 38.7 | #8070 |
| 10 | LLaVA-v1.5-7B | 37.96 | #7933 |
| 11 | Emu2_chat Chat | 37.59 | #7446 |
| 12 | Qwen-VL-Chat | 36.96 | #8089 |
| 13 | mPLUG-Owl2 | 36.21 | #8207 |
| 14 | CogVLM-17B-Chat | 36.08 | #8696 |
| 15 | Yi-VL-6B | 34.31 | #8065 |
No result here: #1 Claude Fable 5.2, #2 GPT-6.2 Astra, #3 Gemini 4 Carbon.
Interactive version: theaggregate.ai/benchmark?slug=gmai-mmbench · How It Works · Data refreshed daily, snapshot 2026-11-17.