Diagnosis on DiagnosisArena
36.36Top-1 AccuracyMixed-Vendor MAC
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Mixed-Vendor MACComposition=Mixed (OpenAI, Gemini, Claude)2026.02 | 36.36 | 49.09 | |
| Single-Vendor MACModel=o4-mini (OpenAI)2026.02 | 35.76 | 47.88 | |
| Single-Vendor MACModel=Gemini-2.5-Pro2026.02 | 33.94 | 44.24 | |
| Single-Vendor MACModel=Claude-4.5-Sonnet2026.02 | 32.73 | 45.45 | |
| Single-LLMModel=o4-mini (OpenAI)2026.02 | 32.12 | 46.06 | |
| Single-LLMModel=Gemini-2.5-Pro2026.02 | 20 | 31.51 | |
| Single-LLMModel=Claude-4.5-Sonnet2026.02 | 19.39 | 29.7 |