MedBrowseComp 121: leaderboard
MedBrowseComp 121-question medical browsing-comprehension split, measuring tool-assisted clinical web-research accuracy.
Metric: Accuracy (%). Source: github.com. 7 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Sonar Pro | 26.12 | #269 |
| 2 | Gemini 2.0 Flash | 18.68 | #342 |
| 3 | GPT-4.1 | 17.36 | #248 |
Interactive version: theaggregate.ai/benchmark?slug=medbrowsecomp-121 · How It Works · Data refreshed daily, snapshot 2026-10-20.