EHRBench - MIMIC-IV: 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-IV records; higher is better.

Metric: Accuracy (%) on the questions built from MIMIC-IV 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

#ModelScoreOverall rank
1GPT-5.2 Instant71.06#309
2GPT-4.169.77#506
3GPT-5 Chat69.18#452
4Llama 3.3 70B Instruct67.94#1394
5GPT-4.1 Mini67.36#799
6Qwen 3 32B67.18#983
7Qwen 2.5 32B67#1347
8GLM-4 32B (0414)66.45#1525
9GPT-5 Mini66.05#373
10m1-32B-1k65.49#1247
11Mistral Small 365.36#1809
12EHRBench llama3-70B (checkpoint unspecified)63.72
13doctor-r1-8B62.26#1634
14Qwen 3 8B62.11#1890
15GPT-4.1 Nano61.89#2065

No result here: #2 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=ehrbench-mimic-iv · How It Works · Data refreshed daily, snapshot 2026-11-22.