Mathematical Reasoning on AIME25 (Accuracy %)
84.17AccuracyParaManager
Evaluation results
| Method | Links | |
|---|---|---|
| ParaManagerBackbone=ParaManager-4B, Orchestration Strategy=Unified, Sampling Protocol (mean@8)=true2026.04 | 84.17 | |
| GPT-OSS-20BOrchestration Strategy=Majority Vote, Sampling Protocol (mean@8)=true2026.04 | 80 | |
| GPT-OSS-20BOrchestration Strategy=Self Refine, Sampling Protocol (mean@8)=true2026.04 | 79.17 | |
| EvoFlowOrchestration Strategy=Static Workflow, Sampling Protocol (mean@8)=true2026.04 | 79.17 | |
| ToolOrchestraOrchestration Strategy=Serial Orchestration, Sampling Protocol (mean@8)=true2026.04 | 79.17 | |
| ParaManager-SerialBackbone=ParaManager-4B, Orchestration Strategy=Serial, Sampling Protocol (mean@8)=true2026.04 | 76.67 | |
| Meta Agent SearchBackbone=Qwen3-30B-A3B-Instruct-2507, Orchestration Strategy=Static Workflow, Sampling Protocol (mean@8)=true2026.04 | 72.92 | |
| RouterOrchestration Strategy=Static Workflow, Sampling Protocol (mean@8)=true2026.04 | 71.25 | |
| GPT-OSS-20BOrchestration Strategy=Base, Sampling Protocol (mean@8)=true2026.04 | 70.42 | |
| GPT-OSS-20BOrchestration Strategy=Base+Tool, Sampling Protocol (mean@8)=true2026.04 | 70.42 | |
| PuppeteerOrchestration Strategy=Serial Orchestration, Sampling Protocol (mean@8)=true2026.04 | 70 | |
| ParaManager-MonoBackbone=ParaManager-4B, Orchestration Strategy=Mono, Sampling Protocol (mean@8)=true2026.04 | 70 | |
| Qwen3-30B A3B-Instruct-2507Orchestration Strategy=Self Refine, Sampling Protocol (mean@8)=true2026.04 | 67.5 | |
| Qwen3-30B A3B-Instruct-2507Orchestration Strategy=Majority Vote, Sampling Protocol (mean@8)=true2026.04 | 66.67 | |
| ParaManager-SFTBackbone=ParaManager-4B, Orchestration Strategy=SFT, Sampling Protocol (mean@8)=true2026.04 | 66.67 | |
| Qwen3-30B A3B-Instruct-2507Orchestration Strategy=Base, Sampling Protocol (mean@8)=true2026.04 | 63.33 | |
| Qwen3-4B Instruct-2507Orchestration Strategy=Self Refine, Sampling Protocol (mean@8)=true2026.04 | 60 | |
| Qwen3-30B A3B-Instruct-2507Orchestration Strategy=Base+Tool, Sampling Protocol (mean@8)=true2026.04 | 58.33 | |
| Qwen3-4B Instruct-2507Orchestration Strategy=Majority Vote, Sampling Protocol (mean@8)=true2026.04 | 56.67 | |
| Qwen3-4B Instruct-2507Orchestration Strategy=Base+Tool, Sampling Protocol (mean@8)=true2026.04 | 52.5 | |
| Qwen3-4B Instruct-2507Orchestration Strategy=Base, Sampling Protocol (mean@8)=true2026.04 | 51.67 | |
| Qwen3-Coder 30B-A3B-InstructOrchestration Strategy=Base+Tool, Sampling Protocol (mean@8)=true2026.04 | 33.33 | |
| Qwen3-Coder 30B-A3B-InstructOrchestration Strategy=Majority Vote, Sampling Protocol (mean@8)=true2026.04 | 26.67 | |
| Qwen3-Coder 30B-A3B-InstructOrchestration Strategy=Self Refine, Sampling Protocol (mean@8)=true2026.04 | 26.67 | |
| Qwen3-Coder 30B-A3B-InstructOrchestration Strategy=Base, Sampling Protocol (mean@8)=true2026.04 | 21.67 |