ClinicBench — leaderboard
Clinical NLP benchmark: evaluates LLMs across multiple clinical tasks including diagnosis, treatment planning, and medical document understanding.
Metric: Average Score. Source: huggingface.co. Status: years away from saturation. 23 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-4 | 56.96 |
| 2 | Llama 3 70B | 52.7 |
| 3 | Claude 2 | 46.35 |
| 4 | GPT-3.5 Turbo | 45.06 |
| 5 | Llama 2 70B | 39.7 |
| 6 | vicuna-13B | 36.48 |
| 7 | Llama 2 13B | 36.32 |
| 8 | Vicuna-7B | 34.61 |
| 9 | Llama 2 7B | 33.68 |
Interactive version: theaggregate.ai/benchmark?slug=clinicbench · How the rankings work · Data refreshed daily, snapshot 2026-07-22.