Mathematical Reasoning on MATH (test) (Cost, Time, Accuracy)
0.0006Execution CostCoT
Evaluation Results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| CoT2026.05 | 0.0006 | — | — | — | — | — | 62.9 | |
| Self-Refine2026.05 | 0.001 | — | — | — | — | — | 65.5 | |
| SC-CoT2026.05 | 0.0037 | — | — | — | — | — | 65.3 | |
| AgentSlimming2026.05 | 0.0069 | — | — | — | — | — | 73.9 | |
| ADAS2026.05 | 0.0115 | — | — | — | — | — | 69 | |
| AFlow2026.05 | 0.012 | — | — | — | — | — | 74.8 | |
| LLM-Debate2026.05 | 0.0147 | — | — | — | — | — | 74.5 | |
| MaASBase Model=GPT-4o-mini2026.04 | 0.42 | 3.38 | 53 | 19 | 3.8 | 72 | 51.82 | |
| GPTSwarmBase Model=GPT-4o-mini2026.04 | 0.93 | 7.02 | 129 | 30 | 7.95 | 159 | 47.88 | |
| AFlowBase Model=GPT-4o-mini2026.04 | 1.66 | 22.5 | 184 | 23 | 24.16 | 207 | 51.28 | |
| SWIFTBase Model=GPT-4o-mini2026.04 | 1.73 | 0.004 | 0.13 | 7 | 1.73 | 7 | 55.14 | |
| MacNetBase Model=GPT-4o-mini2026.04 | 2.35 | — | — | 47 | 2.35 | 47 | 45.18 | |
| DyLANBase Model=GPT-4o-mini2026.04 | 2.89 | 13.01 | 508 | 39 | 15.9 | 547 | 48.63 | |
| LLM-DebateBase Model=GPT-4o-mini2026.04 | 6.76 | — | — | 92 | 6.76 | 92 | 48.54 |