Agentic medical interaction on AgentClinic-MIMIC OOD
50.6Accuracyo3
Evaluation Results
| Method | Links | |
|---|---|---|
| o3Model category=Large Models, Patient agent=GPT-4o2025.07 | 50.6 | |
| Gemini 2.5 ProModel category=Large Models, Patient agent=GPT-4o2025.07 | 48.4 | |
| MedGemma 27BModel category=Small Models, Patient agent=GPT-4o2025.07 | 46 | |
| DeepSeek R1Model category=Large Models, Patient agent=GPT-4o2025.07 | 43.8 | |
| Gemma 3 27BModel category=Small Models, Patient agent=GPT-4o2025.07 | 35.2 |