LMArena.ai (previously known as Chatbot Arena) is an open benchmarking platform that assesses and compares large language models using anonymized, crowd-sourced pairwise evaluations and public voting.
Comparison of Anonymous ModelsAllows users to view and evaluate two AI models side by side without revealing their identities until after voting, guaranteeing an unbiased assessment.
Crowd-sourced voting systemCollects user votes and feedback to produce detailed performance metrics and rankings for various AI models.
All-inclusive Ranking SystemProvides comprehensive performance metrics and rankings derived from more than 3.5 million user votes and various assessment factors.
Testing with Multiple ModalitiesSupports assessment of multiple AI features such as text, visual, and image editing capabilities
