Medical Reasoning on MedBullets
80.8AccuracyMDAgents
Evaluation Results
| Method | Links | |
|---|---|---|
| MDAgentsBase Model=GPT-4o2025.08 | 80.8 | |
| TMA-AllComponBase Model=GPT-4o2025.08 | 78.8 | |
| ReConcileBase Model=GPT-4o2025.08 | 75.2 | |
| MedAgentsBase Model=GPT-4o2025.08 | 70 | |
| DyLANBase Model=GPT-4o2025.08 | 69.5 | |
| TMA-AllComponBase Model=MedGemma-4B2025.08 | 48.2 | |
| TMA-AllComponBase Model=Gemma-3-4B2025.08 | 46.7 | |
| DyLanBase Model=Gemma-3-4B2025.08 | 35.3 | |
| Single-Agent BestBase Model=Gemma-3-4B2025.08 | 34.6 | |
| TMA-AllComponBase Model=Gemma-3-4B2025.08 | 33.9 | |
| MedAgentsBase Model=Gemma-3-4B2025.08 | 33.3 | |
| ReConcileBase Model=Gemma-3-4B2025.08 | 29.3 | |
| MDAgentsBase Model=Gemma-3-4B2025.08 | 22.7 |