Medical Question Answering on Medbullets op4
95.78Pass@1MedXIAOHE
Evaluation Results
| Method | Links | |
|---|---|---|
| MedXIAOHEMode=Thinking mode, Decoding=Greedy2026.02 | 95.78 | |
| Gemini 3.0 ProDecoding=Greedy2026.02 | 91.88 | |
| GPT-5.2 ThinkingMode=Thinking mode, Decoding=Greedy2026.02 | 91.23 | |
| Gemini 2.5 ProDecoding=Greedy2026.02 | 82.79 |