lmring – Open-source platform to compare and self-host AI models

Summary: LMRing is an open-source AI model comparison tool that supports real-time streaming and side-by-side evaluation of up to five models. It enables live rendering of AI-generated code, video output comparison, and integrates over 50 providers with custom API support. The platform features an ELO-based leaderboard driven by user votes and is fully self-hostable to keep data private.

What it does

LMRing allows users to compare multiple AI models simultaneously with real-time streaming, including web development code rendered live and video output comparisons. It supports numerous providers and custom endpoints while maintaining an ELO leaderboard from community voting.

Who it's for

It is designed for developers and AI evaluators who need transparent, extensible, and self-hosted tools to assess various AI models side by side.

Why it matters

LMRing addresses the lack of open, extensible, and multi-modal AI comparison platforms that allow users to bring their own models and retain control over their data.