Mathematical Reasoning on HMMT Nov 25 (Accuracy and Cost)
81.5AccuracyQwen3-235B
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Qwen3-235BMethod Category=Heuristic, Parameters=235B2026.06 | 81.5 | 0.569 | |
| ENTROROUTERMethod Category=Routing Architecture2026.06 | 80 | 0.4163 | |
| Qwen3-30BMethod Category=Heuristic, Parameters=30B2026.06 | 71.7 | 0.2488 | |
| FORCMethod Category=Existing Routing Architecture2026.06 | 70.4 | 0.2731 | |
| RandomMethod Category=Heuristic2026.06 | 68.8 | 0.2783 | |
| Qwen3-4BMethod Category=Heuristic, Parameters=4B2026.06 | 59 | 0.0198 | |
| RouteLLMMethod Category=Existing Routing Architecture2026.06 | 59 | 0.0559 | |
| xRouterMethod Category=Existing Routing Architecture2026.06 | 28.8 | 0.0041 | |
| Router-R1Method Category=Existing Routing Architecture2026.06 | 21.3 | 0.0108 |