Mathematical Reasoning on AIME 24 (Accuracy and Cost)
93.3AccuracyQwen3-235B
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Qwen3-235BMethod Category=Heuristic, Parameters=235B2026.06 | 93.3 | 0.4553 | |
| ENTROROUTERMethod Category=Routing Architecture2026.06 | 90.6 | 0.2881 | |
| FORCMethod Category=Existing Routing Architecture2026.06 | 90 | 0.3435 | |
| Qwen3-30BMethod Category=Heuristic, Parameters=30B2026.06 | 87.5 | 0.2126 | |
| RandomMethod Category=Heuristic2026.06 | 84.6 | 0.2179 | |
| RouteLLMMethod Category=Existing Routing Architecture2026.06 | 80.4 | 0.0433 | |
| Qwen3-4BMethod Category=Heuristic, Parameters=4B2026.06 | 76.9 | 0.0173 | |
| Router-R1Method Category=Existing Routing Architecture2026.06 | 41.7 | 0.0198 | |
| xRouterMethod Category=Existing Routing Architecture2026.06 | 41.7 | 0.0018 |