MR-Bench: leaderboard
MR-Bench (2026) tests decision-oriented clinical reasoning on 1,000 MIMIC-IV hospital admissions with eight-option questions, at temperature 0.5. This board is the mean accuracy of its medication-imputation and procedure-selection tasks, in percent; higher is better.
Metric: Accuracy (%, times 100), mean of the medication-imputation and procedure-selection tasks; 1,000 MIMIC-IV hospital admissions, eight-option questions, temperature 0.5, an answer counted from the first or the last option letter the output names, whichever scores higher over the dataset; higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 24 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Setting | Score | Overall rank |
|---|---|---|---|---|
| 1 | GPT-5 | 64.1 | #178 | |
| 2 | Gemini 3 Pro | 61.3 | #145 | |
| 3 | DeepSeek V3.2 | Thinking | 58.9 | #419 |
| 4 | Qwen 3 Max | 56.4 | #415 | |
| 5 | GPT-4o | 50.6 | #751 | |
| 6 | Qwen 3 8B | 43.6 | #1873 | |
| 7 | MedS3-8B | 43.3 | ||
| 8 | Qwen 3 4B 2507 (Thinking) | 43 | #1433 | |
| 9 | Med42-v2-8B | 41.7 | #2558 | |
| 10 | HuatuoGPT-o1-7B | 41 | #2282 | |
| 11 | Llama 3.1 8B UltraMedical | 40.9 | ||
| 12 | Qwen 2.5 7B Instruct | 40.4 | #2723 | |
| 13 | Adathink-Med-Llama-8B | 40 | ||
| 14 | BioMed-R1-8B | 39.7 | ||
| 15 | m1-7B-23K | 39.1 | #5309 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=mr-bench · How It Works · Data refreshed daily, snapshot 2026-11-21.