# LLM Stats

> Public leaderboards and benchmark site that publishes verifiable evaluations, scores, and performance metrics for large language models and AI providers.

LLM Stats publishes objective leaderboards and benchmark results to show measured model performance rather than marketing claims. The site collects, runs, and displays benchmark results across multiple arenas and datasets, and provides tools to compare models and explore detailed metrics. LLM Stats also offers documentation and an API for programmatic access to results and benchmarks.

- **Leaderboards** — Browse ranked model leaderboards and see comparative scores across benchmarks and arenas.
- **Benchmarks & Arenas** — Access curated benchmark suites (MMLU, GPQA, AIME, etc.) and arena results that evaluate models on domain-specific tasks.
- **Model comparison** — Use the compare tool to view side-by-side performance and metric breakdowns for selected models.
- **Playground & API** — Use the public playground and consult API documentation to programmatically retrieve benchmark data and model metadata.
- **Community & resources** — Read blog posts, community posts, and resources about benchmarks and evaluation methodology.

To get started, visit the website to view leaderboards or benchmark pages, use the compare tool to explore differences between models, and consult the documentation to access the API and playground for automated queries.

## Features
- LLM leaderboards for model rankings
- Curated benchmark suites and arena results
- Model comparison tool
- Public playground for interactive exploration
- API documentation and programmatic access
- News, blog, and community posts about benchmarks

## Integrations
OpenAI, Google, Anthropic, xAI, Alibaba Cloud / Qwen Team, ZeroEval LLM Gateway API

## Platforms
WEB, API

## Pricing
Free

## Links
- Website: https://llm-stats.com/
- Documentation: https://docs.zeroeval.com/
- EveryDev.ai: https://www.everydev.ai/tools/llm-stats