# orq-arena

> LLM arena benchmark on your own prompts: pairwise jury judging, Bradley-Terry ELO with confidence intervals, every token through the orq.ai router.

This is the curated docs index. For the complete docs as one Markdown file, use
[llms-full.txt](llms-full.txt).


## Getting started

- [Home](https://orq-ai.github.io/orq-arena/index.md): Docs overview and suggested reading order.
- [Getting Started](https://orq-ai.github.io/orq-arena/getting-started/index.md): Install, run the zero-key demo, first live tournament.

## Reference

- [CLI Reference](https://orq-ai.github.io/orq-arena/cli/index.md): Every command and flag.
- [Configuration](https://orq-ai.github.io/orq-arena/configuration/index.md): Every orq_arena.yaml key, reasoning recipes, defaults.

## Trust the number

- [Methodology](https://orq-ai.github.io/orq-arena/methodology/index.md): How the ranking is made, bias controls, confidence intervals, reproducibility.