Med-R2 - Anatomy Level: leaderboard
Accuracy (%) of medical vision-language models on the Med-R2 (2026) anatomy level task, locating and identifying anatomy, with equal numbers of positive, negative and neutral adversarial question forms; higher is better.
Metric: Accuracy (%) on the Med-R2 Anatomy Level task (locating and identifying anatomy), zero-shot with system prompts; each task mixes equal numbers of positive, negative and neutral adversarial question forms; higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 14 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Setting | Score | Overall rank |
|---|---|---|---|---|
| 1 | Hulu-Med-14B | 68.1 | #1621 | |
| 2 | HuaTuoGPT-Vision-34B | 65.9 | #1840 | |
| 3 | GPT-5.2 | Thinking | 65.6 | #190 |
| 4 | Qwen 3 VL 235B A22B | 64.2 | #486 | |
| 5 | Qwen 3 VL 8B | 63.3 | #927 | |
| 6 | InternVL3.5-8B | 61 | #1618 | |
| 7 | HuluMed-7B | 60.7 | #2061 | |
| 8 | Hulu-Med-4B | 60.3 | #2410 | |
| 9 | HuatuoGPT-Vision-7B | 58.7 | #2178 | |
| 10 | GPT-4o | 56.1 | #754 | |
| 11 | Lingshu-7B | 56 | #1646 | |
| 12 | Qwen 2.5 VL 7B Instruct | 51.5 | #1919 | |
| 13 | Janus-Pro-7B | 51.3 | #6889 | |
| 14 | MedGemma-4B | 36.4 | #2579 |
No result here: #2 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=med-r2-anatomy-level · How It Works · Data refreshed daily, snapshot 2026-11-22.