lmring

lmring is an open-source, self-hostable platform that lets users run their own LLM arena, enabling side-by-side model comparisons, voting, and automated leaderboards. It’s designed for developers, researchers, and AI enthusiasts who want to evaluate and benchmark custom or private models without relying on external services. Its interest lies in providing full control over evaluation workflows, fostering community-driven model ranking, and enabling transparent, reproducible comparisons in a self-hosted environment.