Math problem solving on Math Macro-aggregate
79.2Pass@1Uno-Orchestra
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Uno-Orchestratok=411, USD/q=0.10112026.05 | 79.2 | 84.7 | |
| Uno-tree-GRPOtok=419, USD/q=0.17122026.05 | 78.5 | 84 | |
| Uno-GRPOtok=391, USD/q=0.16102026.05 | 77.6 | 83 | |
| AOrchestratok=1600, USD/q=0.99322026.05 | 59.3 | 67.1 | |
| ATLAS (cluster)tok=662, USD/q=0.40862026.05 | 56.1 | 63.8 | |
| ICL-Routertok=525, USD/q=0.21002026.05 | 55.7 | 63.5 | |
| xRoutertok=584, USD/q=0.24902026.05 | 55.7 | 63.3 | |
| SkillOrchestratok=533, USD/q=0.18702026.05 | 53.6 | 61.1 | |
| AgentOrchestratok=1724, USD/q=1.21182026.05 | 53.1 | 61.1 | |
| ATLAS (RL)tok=736, USD/q=0.50322026.05 | 52.1 | 60.7 | |
| MasRoutertok=610, USD/q=0.31212026.05 | 50.9 | 58.5 | |
| RouterDCtok=529, USD/q=0.21462026.05 | 48.2 | 55.4 | |
| R2-Reasonertok=601, USD/q=0.35642026.05 | 47.6 | 55.1 | |
| Puppeteertok=694, USD/q=0.46262026.05 | 47.6 | 55.1 | |
| GraphRoutertok=585, USD/q=0.30842026.05 | 45.1 | 52.5 | |
| WideSeek-R1tok=860, USD/q=0.64852026.05 | 41.2 | 48.7 | |
| PromptLLMtok=747, USD/q=0.90712026.05 | 40.2 | 47.2 | |
| Router-R1tok=380, USD/q=0.00052026.05 | 38.9 | 44.5 | |
| ColdStart-LLMtok=540, USD/q=0.22882026.05 | 38.6 | 45.8 | |
| Tool Orchestratok=1896, USD/q=0.81002026.05 | 37.6 | 42.5 | |
| AutoMixtok=1250, USD/q=0.29552026.05 | 34.6 | 40.8 | |
| ToolLLMtok=667, USD/q=0.13552026.05 | 22.7 | 28.3 |