EHRBench - MIMIC-III: leaderboard
EHRBench (2026) tests LLMs on clinical decision questions generated from real patient records and verified against biomedical knowledge bases, posed as multiple-choice questions with 4 to 6 options. This board is accuracy in percent on the questions built from MIMIC-III records; higher is better.
Metric: Accuracy (%) on the questions built from MIMIC-III records, EHR-grounded multiple-choice clinical decision questions (4, 5 and 6 options) built from MIMIC-III, MIMIC-IV and PROMOTE encounter trajectories with knowledge-base verification, JSON-constrained answers, deterministic decoding; higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 31 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.2 Instant | 71.5 | #304 |
| 2 | GPT-4.1 | 70.59 | #505 |
| 3 | GPT-5 Chat | 70.16 | #449 |
| 4 | GPT-5 Mini | 69.08 | #364 |
| 5 | Llama 3.3 70B Instruct | 68.74 | #1385 |
| 6 | GPT-4.1 Mini | 68.58 | #789 |
| 7 | Qwen 3 32B | 68.48 | #972 |
| 8 | Mistral Small 3 | 67 | #1796 |
| 9 | GLM-4 32B (0414) | 66.27 | #1505 |
| 10 | EHRBench llama3-70B (checkpoint unspecified) | 65.39 | |
| 11 | Qwen 2.5 32B | 65.22 | #1344 |
| 12 | m1-32B-1k | 62.87 | #1240 |
| 13 | Qwen 3 8B | 62.26 | #1873 |
| 14 | Qwen 3 4B | 62.01 | #2348 |
| 15 | doctor-r1-8B | 61.94 | #1624 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=ehrbench-mimic-iii · How It Works · Data refreshed daily, snapshot 2026-11-21.