LiveCodeBench
A holistic, contamination-free benchmark that continuously collects new competitive-programming problems from LeetCode, AtCoder, and Codeforces (released after model training cutoffs) and measures code-generation correctness via Pass@1.
What this benchmark measures
A holistic, contamination-free benchmark that continuously collects new competitive-programming problems from LeetCode, AtCoder, and Codeforces (released after model training cutoffs) and measures code-generation correctness via Pass@1.
Rows on this page are sourced from public benchmark artifacts, leaderboard exports, or source-linked model reports. Each row keeps benchmark version, source model name, and available run details attached to the score.
The metric shown here is Pass@1. It should be interpreted within LiveCodeBench, not compared as part of a site-wide ranking.
Frequently asked
What is LiveCodeBench?
A holistic, contamination-free benchmark that continuously collects new competitive-programming problems from LeetCode, AtCoder, and Codeforces (released after model training cutoffs) and measures code-generation correctness via Pass@1. It is a coding benchmark measured by Pass@1.
What does Pass@1 mean on LiveCodeBench?
LiveCodeBench reports Pass@1 (%); higher is better. Scores are shown only within LiveCodeBench and are never averaged with other benchmarks.
What is the top reported LiveCodeBench score?
Fugu Ultra has the top reported score on LiveCodeBench: 93.2% (Pass@1).
Why do LiveCodeBench scores differ across runs?
Harness, scaffold, reasoning effort, and prompt setup change results, so two runs of the same model can differ. evals.report keeps each score with its run context so the differences stay visible.
Does evals.report rank models across benchmarks?
No. LiveCodeBench scores are shown within their own metric; evals.report never combines benchmarks into a composite ranking or a single "best model".