Medical Question Answering on CMExam (Pass@1)
96.12Pass@1MedXIAOHE
Evaluation results
| Method | Links | |
|---|---|---|
| MedXIAOHEMode=Thinking mode, Decoding=Greedy2026.02 | 96.12 | |
| Gemini 3.0 ProDecoding=Greedy2026.02 | 93.13 | |
| GPT-5.2 ThinkingMode=Thinking mode, Decoding=Greedy2026.02 | 89.12 | |
| Gemini 2.5 ProDecoding=Greedy2026.02 | 87.49 |