BigCodeBench

BigCodeBench The dataset has 2 variants: BigCodeBench-Complete: Code Completion based on the structured docstrings. BigCodeBench-Instruct: Code Generation based on the NL-oriented instructions. The overall statistics of the dataset are as follows: Complete Instruct # Task 1140 1140 # Avg. Test Cases 5.6 5.6 # Avg. Cove

Papers using BigCodeBench (3)

Conv-to-Bench: Evaluating Language Models Via User-Assistant Dialogues In Code Tasks2026ReflexiCoder: Teaching Large Language Models to Self-Reflect on Generated Code and Self-Correct It via Reinforcement Learning2026OpenCodeInstruct: A Large-scale Instruction Tuning Dataset for Code LLMs2025