Code Generation on MBPP-ET

Metric: Pass@1 (higher is better)

LeaderboardDataset
Loading chart...
#ModelPass@1Extra DataPaperDateCode
1EG-CFG (DeepSeek-V3-0324)73NoExecution Guided Line-by-Line Code Generation2025-06-12Code
2LPW (GPT-4o)65.8NoPlanning-Driven Programming: A Large Language Mo...2024-11-21Code