EHRBench: leaderboard
Clinical decision-making (CDM) is central to real-world clinical workflows, where clinicians infer diagnoses, select treatments, or anticipate future health outcomes under incomplete evidence. LLMs are increasingly used to support these decisions due to strong language capabilities, broad biomedical knowledge, and efficiency, yet the reliability of LLMs on real-world clinical decision tasks remain
Metric: Overall Acc (%) (self-reported). Source: benchmarklist.com. Saturation forecast: Around January 2027. 31 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.2 | 70.91 | #110 |
| 2 | GPT-4.1 | 69.43 | #248 |
| 3 | GPT-5 | 69.06 | #98 |
| 4 | Llama 3.3 70B | 67.28 | #579 |
| 5 | GPT-4.1 Mini | 66.79 | #355 |
| 6 | Qwen 3 32B | 66.78 | #443 |
| 7 | GPT-5 Mini | 66.12 | #186 |
| 8 | GLM-4 32B | 66.12 | #459 |
| 9 | Qwen 2.5 32B | 64.97 | #546 |
| 10 | Llama 3 70B | 63.35 | #631 |
| 11 | Qwen 3 8B | 60.87 | #674 |
| 12 | Qwen 3 4B | 60.63 | #829 |
| 13 | GPT-4.1 Nano | 60.48 | #731 |
| 14 | glm-4-9B | 59.62 | #889 |
| 15 | Yi 1.5 34B | 58.94 | #807 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=ehrbench · How It Works · Data refreshed daily, snapshot 2026-10-20.