MMRareBench - Treatment Planning: leaderboard
MMRareBench (2026) treatment-planning track: given the working diagnosis, case narrative and medical images of a rare-disease case, the model writes a stage-wise plan covering safety, monitoring and follow-up. Rubric score (0-100) from a strict Qwen3-VL-235B judge; higher is better.
Metric: Treatment Planning track (T2) score: a stage-wise treatment plan given the working diagnosis, case narrative and images, graded 0-100 by a Qwen3-VL-235B judge with a strict track-specific rubric that awards credit only when a criterion is fully met (six binary dimensions averaged); higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 23 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5 | 49.2 | #178 |
| 2 | Qwen 3 VL 235B A22B Instruct | 35.9 | #532 |
| 3 | Claude Sonnet 4.5 | 35.8 | #261 |
| 4 | Claude Haiku 4.5 | 34.2 | #583 |
| 5 | Gemini 3 Flash (Preview) | 29.1 | #152 |
| 6 | Gemini 2.5 Pro | 26.7 | #276 |
| 7 | GLM-4.6V | 20.5 | #710 |
| 8 | Qwen 3 VL 30B A3B Instruct | 20.1 | #895 |
| 9 | Gemini 2.5 Flash | 17.1 | #463 |
| 10 | MedGemma-27B-IT | 15.4 | #1350 |
| 11 | Lingshu-32B | 14.2 | #1113 |
| 12 | Hulu-Med-14B | 14.2 | #1607 |
| 13 | HealthGPT-L14B | 14.1 | #1397 |
| 14 | Qwen 2.5 VL 32B Instruct | 12.8 | #1062 |
| 15 | GPT-4o | 12.7 | #751 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=mmrarebench-treatment-planning · How It Works · Data refreshed daily, snapshot 2026-11-21.