Knowledge and reasoning on MedXpertQA text + MM
74.4Accuracy (MedXpertQA Text+MM)Gemini 3 Pro
Evaluation Results
| Method | Links | |
|---|---|---|
| Gemini 3 ProTemperature=default2026.04 | 74.4 | |
| Gemini 3 FlashTemperature=default2026.04 | 72.8 | |
| MedGemma 1Model Scale=27B, Temperature=0.0, OOD Status=out of distribution2026.04 | 26.8 | |
| MedGemma 1.5Model Scale=4B, Temperature=0.0, OOD Status=out of distribution2026.04 | 26.4 | |
| Qwen3 VLModel Scale=4B, Temperature=default2026.04 | 21.9 | |
| MedGemma 1Model Scale=4B, Temperature=0.0, OOD Status=out of distribution2026.04 | 18.8 |