MedAgentBench: leaderboard
Interactive EHR-agent benchmark with physician-written tasks over healthcare data and FHIR-style clinical workflows.
Metric: Overall SR (imported). Source: stanfordmlgroup.github.io. Source: benchmarklist.com. Saturation forecast: Estimated already saturated. 12 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-4o | 64 | #343 |
| 2 | DeepSeek V3 | 62.67 | #321 |
| 3 | Gemini 1.5 Pro | 62 | #445 |
| 4 | GPT-4o Mini | 56.33 | #599 |
| 5 | O3 Mini | 51.67 | #272 |
| 6 | Llama 3.3 70B | 46.33 | #579 |
| 7 | Gemini 2.0 Flash | 38.33 | #342 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=medagentbench · How It Works · Data refreshed daily, snapshot 2026-10-20.