LiveCodeBench v-5

LiveCodeBench v5 is a benchmark dataset that contains a collection of synthetic programming tasks, solutions, and test cases used to evaluate the performance of code reasoning models in competitive programming scenarios.

Papers using LiveCodeBench v-5 (7)

X-Coder: Advancing Competitive Programming with Fully Synthetic Tasks, Solutions, and Tests2026Critique-Coder: Enhancing Coder Models by Critique Reinforcement Learning2025PromptCoT 2.0: Scaling Prompt Synthesis for Large Language Model Reasoning2025PromptCoT 2.0: Scaling Prompt Synthesis for Large Language Model Reasoning2025Critique-Coder: Enhancing Coder Models by Critique Reinforcement Learning2025Pruning the Unsurprising: Efficient Code Reasoning via First-Token Surprisal2025X-Coder: Advancing Competitive Programming with Fully Synthetic Tasks, Solutions, and Tests2026