List your company

Mathematical Reasoning on MATH (test) (Cost, Time, Accuracy)

0.0006Execution CostCoT
-0.2697731.5552643.3803015.205339Apr 27, 2026Apr 29, 2026May 1, 2026May 3, 2026May 5, 2026May 7, 2026May 9, 2026

Evaluation results

MethodLinks
CoT2026.050.0006—————62.9
Self-Refine2026.050.001—————65.5
SC-CoT2026.050.0037—————65.3
AgentSlimming2026.050.0069—————73.9
ADAS2026.050.0115—————69
AFlow2026.050.012—————74.8
LLM-Debate2026.050.0147—————74.5
MaASBase Model=GPT-4o-mini2026.040.423.3853193.87251.82
GPTSwarmBase Model=GPT-4o-mini2026.040.937.02129307.9515947.88
AFlowBase Model=GPT-4o-mini2026.041.6622.51842324.1620751.28
SWIFTBase Model=GPT-4o-mini2026.041.730.0040.1371.73755.14
MacNetBase Model=GPT-4o-mini2026.042.35——472.354745.18
DyLANBase Model=GPT-4o-mini2026.042.8913.015083915.954748.63
LLM-DebateBase Model=GPT-4o-mini2026.046.76——926.769248.54