List your company

Mathematical Reasoning on MATH 500 (Token Usage Metrics)

2,025,345Avg Token Usage (R0)Elim
1,924,077.751,974,711.3752,025,3452,075,978.625Apr 22, 2026

Evaluation results

MethodLinks
ElimModel=GPT-4.1-mini, Seeds=42026.042,025,3451,680,1731,308,1441,249,4196,263,081
Self-Evolving Test-Time AllocationModel=GPT-4.1-mini, Seeds=42026.042,025,3451,102,486902,109741,6114,771,551