List your company

Mathematical Reasoning on AIME 2024 (Avg@32, Pass@8, Dist-N, SemDiv)

50.63Avg@32CraEG
50.411650.468350.52550.5817Jan 30, 2026

Evaluation results

MethodLinks
CraEGModel=Qwen3-4B, Decoding configuration=Temp = 1.0, Top-p = 1.02026.0150.6360.6854.616.3
Standard SamplingModel=Qwen3-4B, Decoding configuration=Temp = 1.0, Top-p = 1.02026.0150.4260.9651.1414.99