Multi-turn Medical Diagnosis on Med-Inquire DiagnosisArena (915 cases)
59.8Mean Diagnostic GradeEvoClinician
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| EvoClinicianBackbone=gemini-3-pro2026.01 | 59.8 | 9.1 | 1,275 | |
| EvoTestBackbone=gemini-3-pro2026.01 | 57.9 | 9.4 | 1,605 | |
| EvoClinicianBackbone=gpt-5.12026.01 | 54.2 | 9.3 | 1,310 | |
| EvoPromptBackbone=gemini-3-pro2026.01 | 53.6 | 9.7 | 1,360 | |
| EvoTestBackbone=gpt-5.12026.01 | 52.5 | 9.6 | 1,650 | |
| Evo-MemoryBackbone=gemini-3-pro2026.01 | 52 | 10 | 1,435 | |
| Mem0Backbone=gemini-3-pro2026.01 | 51.2 | 10.1 | 1,450 | |
| EvoClinicianBackbone=MedGemma-27b2026.01 | 51.1 | 9.5 | 1,328 | |
| RAGBackbone=gemini-3-pro2026.01 | 50.7 | 10.3 | 1,490 | |
| MemoryBackbone=gemini-3-pro2026.01 | 50.1 | 10.4 | 1,510 | |
| EvoTestBackbone=MedGemma-27b2026.01 | 49.6 | 9.7 | 1,670 | |
| GEPABackbone=gemini-3-pro2026.01 | 49.4 | 10.6 | 1,540 | |
| EvoPromptBackbone=gpt-5.12026.01 | 49 | 9.9 | 1,390 | |
| Static PromptBackbone=gemini-3-pro2026.01 | 48.2 | 9.8 | 1,380 | |
| EvoClinicianBackbone=gpt-4o-mini2026.01 | 48 | 9.6 | 1,345 | |
| Evo-MemoryBackbone=gpt-5.12026.01 | 47.1 | 10.3 | 1,465 | |
| EvoPromptBackbone=MedGemma-27b2026.01 | 46.6 | 10 | 1,405 | |
| EvoTestBackbone=gpt-4o-mini2026.01 | 46.6 | 9.8 | 1,690 | |
| Mem0Backbone=gpt-5.12026.01 | 46.3 | 10.4 | 1,480 | |
| RAGBackbone=gpt-5.12026.01 | 45.7 | 10.6 | 1,520 | |
| MemoryBackbone=gpt-5.12026.01 | 45.1 | 10.7 | 1,540 | |
| Evo-MemoryBackbone=MedGemma-27b2026.01 | 44.8 | 10.5 | 1,480 | |
| GEPABackbone=gpt-5.12026.01 | 44.6 | 10.9 | 1,575 | |
| EvoPromptBackbone=gpt-4o-mini2026.01 | 44.1 | 10.2 | 1,420 | |
| Mem0Backbone=MedGemma-27b2026.01 | 44 | 10.6 | 1,495 | |
| Static PromptBackbone=gpt-5.12026.01 | 43.6 | 10.1 | 1,410 | |
| RAGBackbone=MedGemma-27b2026.01 | 43.3 | 10.8 | 1,535 | |
| MemoryBackbone=MedGemma-27b2026.01 | 42.8 | 10.9 | 1,560 | |
| Evo-MemoryBackbone=gpt-4o-mini2026.01 | 42.5 | 10.6 | 1,495 | |
| GEPABackbone=MedGemma-27b2026.01 | 42.2 | 11 | 1,593 | |
| Mem0Backbone=gpt-4o-mini2026.01 | 41.7 | 10.7 | 1,510 | |
| Static PromptBackbone=MedGemma-27b2026.01 | 41 | 10.2 | 1,425 | |
| RAGBackbone=gpt-4o-mini2026.01 | 40.9 | 10.9 | 1,550 | |
| MemoryBackbone=gpt-4o-mini2026.01 | 40.4 | 11 | 1,580 | |
| GEPABackbone=gpt-4o-mini2026.01 | 39.8 | 11.2 | 1,610 | |
| Static PromptBackbone=gpt-4o-mini2026.01 | 38.9 | 10.4 | 1,440 |