List your company

Single-Turn Mathematical Reasoning on MATH500

93.43Peak Avg ScoreOTB
90.382891.173991.96592.7561Feb 6, 2026

Evaluation results

MethodLinks
OTBBackbone=Qwen3-8b-Base, Reasoning Paradigm=Single-Turn2026.0293.43
OPOBackbone=Qwen3-8b-Base, Reasoning Paradigm=Single-Turn2026.0293.31
OGBBackbone=Qwen3-8b-Base, Reasoning Paradigm=Single-Turn2026.0290.75
RLOOBackbone=Qwen3-8b-Base, Reasoning Paradigm=Single-Turn2026.0290.56
GRPOBackbone=Qwen3-8b-Base, Reasoning Paradigm=Single-Turn2026.0290.5