MedS-Bench - MCQA: leaderboard
The MCQA subtask of MedS-Bench (arXiv) — multiple-choice medical question answering; accuracy (%). Published overview values are retained where detailed tables differ.
Metric: Accuracy (%). Source: henrychur.github.io. 5 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-4 | 75.3 | #438 |
Interactive version: theaggregate.ai/benchmark?slug=meds-bench-mcqa · How It Works · Data refreshed daily, snapshot 2026-10-20.