Data as of Jul 25, 2026 · Based on 2,718,867 AI responses across 9,511 prompts · See how Parse measures this
Decentralized Arenagithub.com · Low evidenceDe Arena is a fully automated framework that uses collective intelligence from all LLMs to evaluate each other, providing robust and unbiased rankings. It achieves a 97% correlation with Chatbot Arena on over 60 models while enabling fast, transparent, and customizable evaluation across any number of dimensions.
Parse Score