LiveCodeBench Pro
A live competitive-programming benchmark that rates LLMs with a Codeforces-style Elo on fresh contest problems.
What this benchmark measures
A live competitive-programming benchmark that rates LLMs with a Codeforces-style Elo on fresh contest problems.
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 Codeforces Elo. It should be interpreted within LiveCodeBench Pro, not compared as part of a site-wide ranking.
What to be careful about
Elo is contest-relative and not comparable to pass@1 coding benchmarks; keep it separate.
Frequently asked
What is LiveCodeBench Pro?
A live competitive-programming benchmark that rates LLMs with a Codeforces-style Elo on fresh contest problems. It is a coding benchmark measured by Codeforces Elo.
What does Codeforces Elo mean on LiveCodeBench Pro?
LiveCodeBench Pro reports Codeforces Elo; higher is better. Scores are shown only within LiveCodeBench Pro and are never averaged with other benchmarks.
What is the top reported LiveCodeBench Pro score?
Gemini 3 Deep Think has the top reported score on LiveCodeBench Pro: 3298 (Codeforces Elo).
Why do LiveCodeBench Pro 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 Pro scores are shown within their own metric; evals.report never combines benchmarks into a composite ranking or a single "best model".