List your company

Hard Math Reasoning on AIME 2025 (Pass@1, Pass@8, Delta Pass@1)

13.33Pass@1MIG
-0.53323.06596.66510.2641Feb 1, 2026

Evaluation results

MethodLinks
MIGEvaluation Mode=Out-of-Distribution (OOD), Training Strategy=Transferred from MATH (300 steps)2026.0213.332013.33
Base ModelEvaluation Mode=Out-of-Distribution (OOD), Training Strategy=Transferred from MATH (300 steps)2026.02020—
GRPOEvaluation Mode=Out-of-Distribution (OOD), Training Strategy=Transferred from MATH (300 steps)2026.02013.33—