List your company

Code Generation on LiveCodeBench v6 (Pass@4)

62.9Pass@4CPPO
-2.2341614.6756731.585548.49533May 26, 2026Jun 1, 2026Jun 7, 2026Jun 13, 2026Jun 19, 2026Jun 25, 2026Jul 1, 2026

Evaluation results

MethodLinks
CPPOModel=Gemma 4 E4B, K budget=42026.0562.9
PKPOModel=Gemma 4 E4B, K budget=42026.0560.1
Pass@K Training / RLVRModel=Gemma 4 E4B, K budget=42026.0559.8
CPPOModel=Gemma 4 E2B, K budget=42026.0556.1
PlanSearchModel=Gemma 4 E4B, K budget=42026.0556
Plan-and-SolveModel=Gemma 4 E4B, K budget=42026.0553.1
PKPOModel=Gemma 4 E2B, K budget=42026.0552.1
Direct SolveModel=Gemma 4 E4B, K budget=42026.0552
Pass@K Training / RLVRModel=Gemma 4 E2B, K budget=42026.0551.9
PlanSearchModel=Gemma 4 E2B, K budget=42026.0548.1
Plan-and-SolveModel=Gemma 4 E2B, K budget=42026.0545.2
Direct SolveModel=Gemma 4 E2B, K budget=42026.0544
PMDModel=Qwen3-8B2026.070.517
PMDModel=OLMo3-Instruct-7B2026.070.511
SDPOModel=Qwen3-8B2026.070.479
SDPOModel=OLMo3-Instruct-7B2026.070.45
GRPOModel=Qwen3-8B2026.070.412
GRPOModel=OLMo3-Instruct-7B2026.070.361
Base PolicyModel=OLMo3-Instruct-7B2026.070.277
Base PolicyModel=Qwen3-8B2026.070.271