Medical Reasoning on DDXPlus (Token Cost)
2,383Token CostTMA-AllCompon
Evaluation Results
| Method | Links | |
|---|---|---|
| TMA-AllComponModel=MedGemma-4B, Framework=TMA-AllCompon2025.08 | 2,383 | |
| TMA-AllComponModel=Gemma-3-4B, Framework=TMA-AllCompon2025.08 | 2,383 | |
| TMA-AllComponModel=GPT-4o, Framework=TMA-AllCompon2025.08 | 2,383 | |
| DyLANModel=GPT-4o, Framework=DyLAN2025.08 | 3,271 | |
| ReConcileModel=GPT-4o, Framework=ReConcile2025.08 | 4,456 | |
| DyLANModel=Gemma-3-4B, Framework=DyLAN2025.08 | 6,216 | |
| ReConcileModel=Gemma-3-4B, Framework=ReConcile2025.08 | 8,145 | |
| MDAgentsModel=Gemma-3-4B, Framework=MDAgents2025.08 | 31,206 | |
| MDAgentsModel=GPT-4o, Framework=MDAgents2025.08 | 31,206 | |
| MedAgentsModel=Gemma-3-4B, Framework=MedAgents2025.08 | 41,862 | |
| MedAgentsModel=GPT-4o, Framework=MedAgents2025.08 | 41,862 |