List your company

Code Reasoning on HumanEval+ (avg@10)

41.6Avg Score (k=10)AntiSD
40.35240.6764141.324May 12, 2026

Evaluation results

MethodLinks
AntiSDModel=Qwen3-8B2026.0541.6
GRPOModel=Qwen3-8B2026.0540.4