MedXpertQA Text (Natomy): leaderboard
Accuracy on the MedXpertQA text-only set - 2,450 expert-level medical exam questions with ten answer choices - run zero-shot with chain-of-thought reasoning.
Metric: Accuracy (%). Source: www.natomy.com. Saturation forecast: Around February 2027. 6 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.8 Flash | 66.2 | #30 |
| 2 | GPT-5.6 Sol | 59.63 | #22 |
| 3 | Claude Sonnet 5 | 49.31 | #65 |
| 4 | GPT-4o | 30.37 | #357 |
| 5 | Claude 3.5 Sonnet | 21.31 | #362 |
| 6 | Gemini 2.0 Flash | 20.57 | #351 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=medxpertqa-text-natomy · How It Works · Data refreshed daily, snapshot 2026-11-05.