Chatbot Arena

Free

An open platform for crowdsourced AI benchmarking, hosted by researchers at UC Berkeley SkyLab and LMArena.

FreeFree tier

Type

Open Source

Visit Website GitHub (40 stars)

About Chatbot Arena

Created by researchers from UC Berkeley, Arena (formerly LMArena) is a community-powered platform for understanding AI performance in the real world. Tens of millions of builders, researchers, and creative professionals use Arena to test frontier models and provide feedback on their responses, shaping a public leaderboard grounded in real-world use. The platform's mission is to measure and advance the frontier of AI for real-world use, with a vision to build the foundation for everyone to understand, shape, and benefit from AI. Arena also offers an enterprise AI Evaluations service for comprehensive evaluation services based on real-world human feedback.

Key Features

Crowdsourced AI model evaluation and benchmarking

Public leaderboard grounded in real-world human feedback

Battle mode for comparing models side-by-side

Community of millions of builders, researchers, and professionals

Enterprise AI evaluation services

Pros & Cons

Pros

Open and transparent leaderboard based on real user feedback
Powered by a large community of diverse users
Backed by UC Berkeley academic research
Free to use for individuals and researchers
Supports multiple frontier models for comparison

Cons

Relies on crowdsourced feedback which may have inherent biases
Primarily focused on language model evaluation

Best For

Comparing large language model performanceConducting AI research on model capabilitiesEnterprise evaluation of AI models with real-world feedbackCommunity-driven benchmarking for model development