LiveCodeBench A
AI benchmark · Coding
What is LiveCodeBench?
LiveCodeBench measures LLM coding ability on continuously collected competitive programming problems from LeetCode, AtCoder and CodeForces, covering code generation, self-repair, code execution and test output prediction while avoiding contamination.
Coding mostly
“we propose LiveCodeBench, a comprehensive and contamination-free evaluation of LLMs for code”
arxiv.org/abs/2403.07974 ↗
Paper: arxiv.org/abs/2403.07974 ↗
Lineage
Benchmarks this one is built on and built from: versions, subsets, extensions and the indexes that combine them.
Derived benchmarks
AA LiveCodeBench extension
“AA LiveCodeBench is Artificial Analysis's independently run LiveCodeBench evaluation scoring models' code generation ability.”
CoREB extension
“CoREB measures code retrieval and reranking models across text-to-code, code-to-text, and code-to-code search tasks built from rewritten LiveCodeBench problems in five programming languages.”
LiveCodeBench Pro later version
LiveCodeBench v5 later version
LiveCodeBench v6 later version
LiveCodeBench-Plus extension
Multi-LCB extension
“Multi-LCB measures LLM code-generation ability on contamination-aware competitive programming tasks across twelve programming languages, extending LiveCodeBench beyond Python.”
Tier A
Widely used: a lab reports it, or 50 or more models are scored on it.
How much stands behind a tier: usage, who backs it, a confirmed line from its own site or paper, and a clear read on X. High with three of the four, medium with two, low otherwise.
- 122 models scored
- in curated RL directories (2 pts)
On X · well regarded, moderate buzz, from 7 posts about LiveCodeBench in the last month
LiveCodeBench is routinely cited as a standard coding eval in quant tests, fine-tune claims and agent framework results, though scores near 90% hint at saturation.
How we got to tier A
- Adoption: 122 models scored. Tier A.
- On X: well regarded, moderate buzz. “LiveCodeBench is routinely cited as a standard coding eval in quant tests, fine-tune claims and agent framework results, though scores near 90% hint at saturation.” No move.
- In curated RL directories: 2 points, a floor of C.
- Confirmed: a line from its own site or paper places it.
A clear view on X moves a tier one step up or down; most entries don't move. How tiers work.
Where LiveCodeBench sits
Its domain on the map, live. Hover a logo for its name; click to open it. Open it on the full map
Next: browse all benchmarks. Entry from the RL Research daily scrape of public sources, 2026-11-05 snapshot.