Mathematical Reasoning on MATH 500 (Acc., Cost)
99.2AccuracyQwen3-235B
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Qwen3-235BMethod Category=Heuristic, Parameters=235B2026.06 | 99.2 | 2.5881 | |
| ENTROROUTERMethod Category=Routing Architecture2026.06 | 98.4 | 0.5238 | |
| Qwen3-30BMethod Category=Heuristic, Parameters=30B2026.06 | 98.2 | 1.2123 | |
| FORCMethod Category=Existing Routing Architecture2026.06 | 98.2 | 1.7675 | |
| RandomMethod Category=Heuristic2026.06 | 98 | 1.279 | |
| RouteLLMMethod Category=Existing Routing Architecture2026.06 | 97.6 | 0.1042 | |
| Qwen3-4BMethod Category=Heuristic, Parameters=4B2026.06 | 97.4 | 0.0907 | |
| Router-R1Method Category=Existing Routing Architecture2026.06 | 93.4 | 0.0018 | |
| xRouterMethod Category=Existing Routing Architecture2026.06 | 93.2 | 0.0009 |