LLM Routing on SUPERGPQA
77.6Top-1 AccORACLE
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| ORACLEModel Pool=Pool Large, Training Data Source=Gemini-2.5-Flash2026.01 | 77.6 | — | |
| ORACLEModel Pool=Pool Small, Training Data Source=Gemini-2.5-Flash2026.01 | 67.8 | — | |
| TOP-1 / TOP-3Model Pool=Pool Large, Training Data Source=Gemini-2.5-Flash2026.01 | 54.4 | 54.7 | |
| LLMRANKModel Pool=Pool Large, Router Category=Query-answer Routers, Training Data Source=Gemini-2.5-Flash2026.01 | 52.6 | — | |
| AVENGERSModel Pool=Pool Large, Router Category=Query-answer Routers, Training Data Source=Gemini-2.5-Flash2026.01 | 51.7 | — | |
| CASCAL-GTModel Pool=Pool Large, Router Category=Query-answer Routers, Training Data Source=Gemini-2.5-Flash2026.01 | 51.3 | 53.4 | |
| SMOOTHIEModel Pool=Pool Large, Router Category=Query-only Routers, Training Data Source=Gemini-2.5-Flash2026.01 | 44.5 | 48.5 | |
| CASCALModel Pool=Pool Large, Router Category=Query-only Routers, Training Data Source=Gemini-2.5-Flash2026.01 | 42.9 | 52.3 | |
| TOP-1 / TOP-3Model Pool=Pool Small, Training Data Source=Gemini-2.5-Flash2026.01 | 37.3 | 35.7 | |
| LLMRANKModel Pool=Pool Small, Router Category=Query-answer Routers, Training Data Source=Gemini-2.5-Flash2026.01 | 36.4 | — | |
| CASCAL-GTModel Pool=Pool Small, Router Category=Query-answer Routers, Training Data Source=Gemini-2.5-Flash2026.01 | 36.3 | 31.4 | |
| AVENGERSModel Pool=Pool Small, Router Category=Query-answer Routers, Training Data Source=Gemini-2.5-Flash2026.01 | 36.3 | — | |
| CASCALModel Pool=Pool Small, Router Category=Query-only Routers, Training Data Source=Gemini-2.5-Flash2026.01 | 36.2 | 31.7 | |
| SMOOTHIEModel Pool=Pool Small, Router Category=Query-only Routers, Training Data Source=Gemini-2.5-Flash2026.01 | 31.8 | 32.2 |