LiveBench/LiveBench
LiveBench is an LLM benchmark that releases new evaluation questions monthly to test model capabilities without data contamination.
Not currently ranked — collecting fresh signals.
star history
LiveBench is a benchmark suite for evaluating large language models with contamination-resistant test sets. It releases new questions monthly based on recent datasets, arXiv papers, news articles, and IMDb synopses. Each task includes verifiable, objective ground-truth answers enabling accurate, automated scoring without LLM judges. The framework covers 18 diverse tasks across 6 categories and maintains a public leaderboard for model comparison.
Frequently asked
- What is LiveBench/LiveBench?
- LiveBench is an LLM benchmark that releases new evaluation questions monthly to test model capabilities without data contamination.
- Is LiveBench open source?
- Yes — LiveBench/LiveBench is an open-source project tracked on heatdrop.
- What language is LiveBench written in?
- LiveBench/LiveBench is primarily written in Python.
- How popular is LiveBench?
- LiveBench/LiveBench has 1.3k stars on GitHub.
- Where can I find LiveBench?
- LiveBench/LiveBench is on GitHub at https://github.com/LiveBench/LiveBench.