MedQA
Health & Medicine · 2020-09-28
MedQA evaluates medical question-answering ability on medical exam-style questions. Scores are usually reported as accuracy percentages.
Top models (higher is better)
| Model | Score |
|---|---|
| o3 | 93.3 |
| Gemini 2.5 Pro | 92.6 |
| Gemini 2.5 Flash | 92.0 |
| MedGemma 27B Text IT | 87.7 |
| MedGemma 27B IT | 85.3 |
| Gemma 3 27B IT | 74.9 |
| MedGemma 1.5 4B IT | 69.1 |
| MedGemma 4B IT | 64.4 |
| Gemma 3 4B IT | 50.7 |