List your company

Mathematical Reasoning on Olympiad standard (test)

38.6Pass@1AGPO
24.45628.12831.835.472May 7, 2026

Evaluation results

MethodLinks
AGPOBackbone=Qwen2.5-Math-7B2026.0538.645.451.757.56265.969.472.575.3
GRPOBackbone=Qwen2.5-Math-7B2026.0537.544.250.355.760.263.866.969.572
PPOBackbone=Qwen2.5-Math-7B2026.0537.143.749.855.159.663.266.369.171.9
W-REINFORCEBackbone=Qwen2.5-Math-7B2026.0536.944.250.856.561.365.46972.175.1
REINFORCEBackbone=Qwen2.5-Math-7B2026.0535.142.547.952.357.861.466.769.372.5
Base ModelBackbone=Qwen2.5-Math-7B2026.052535.244.45258.363.467.671.374.4