List your company

Mathematical Reasoning on MATH 500 (Reasoning Performance)

93.4Reasoning PerformanceImproved CoT
89.2490.3291.492.48May 21, 2025

Evaluation results

MethodLinks
Improved CoTModel Series=32B Models, Safety Training Data=10002025.0593.4
RealSafe-R1-7BModel Series=7B Models, Safety Training Data=150002025.0592
RealSafe-R1-32BModel Series=32B Models, Safety Training Data=150002025.0591.6
STAR1-R1-Distill-7BModel Series=7B Models, Safety Training Data=10002025.0590.4
Improved CoTModel Series=7B Models, Safety Training Data=10002025.0590.2
STAR1-R1-Distill-32BModel Series=32B Models, Safety Training Data=10002025.0589.4