MR-Bench - Procedure Selection: 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 asks for the set of procedures performed in the admission, among procedures of other patients. Accuracy in percent; higher is better.
Metric: Accuracy (%, times 100) at picking the admission's procedure set against other patients' procedures; 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 | 73.8 | #178 | |
| 2 | DeepSeek V3.2 | Thinking | 69.4 | #419 |
| 3 | Qwen 3 Max | 68.2 | #415 | |
| 4 | Gemini 3 Pro | 67.6 | #145 | |
| 5 | GPT-4o | 64.9 | #751 | |
| 6 | Qwen 3 8B | 60.2 | #1873 | |
| 7 | HuatuoGPT-o1-7B | 59.9 | #2282 | |
| 8 | Qwen 2.5 7B Instruct | 59.5 | #2723 | |
| 9 | Qwen 3 4B 2507 (Thinking) | 58.8 | #1433 | |
| 10 | Llama 3.1 8B UltraMedical | 58.6 | ||
| 11 | Med42-v2-8B | 58.5 | #2558 | |
| 12 | MedS3-8B | 57.8 | ||
| 13 | m1-7B-23K | 57.5 | #5309 | |
| 14 | Adathink-Med-Llama-8B | 57.4 | ||
| 15 | BioMed-R1-8B | 55.2 |
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-procedure-selection · How It Works · Data refreshed daily, snapshot 2026-11-21.