LMArena

LMArena is a public, transparent evaluation platform for large language models and multimodal AI systems. The platform allows users to input prompts, compare AI-generated answers from multiple models (both open-source and proprietary), and vote on which model produces the best response across a range of tasks—including text, code, image, video, document, and search. Crowd-sourced feedback directly shapes public leaderboards and supports researchers, users, and developers in advancing state-of-the-art AI. LMArena also provides open datasets for academic and community research.