List your company

Code Reasoning on LiveCodeBench (Pass@1, Pass@5, Pass@10)

28.7Pass@1HTPO
24.0225.23526.4527.665May 8, 2026

Evaluation results

MethodLinks
HTPOBase Model=Qwen3-8B2026.0528.734.736.1
DAPOBase Model=Qwen3-8B2026.0526.832.834.3
GRPOBase Model=Qwen3-8B2026.0524.230.231.9