LM Arena
Crowdsourced AI Model Evaluation Platform
Creators mentioned
1
Total mentions
1
Category
Research & search
What it does
Arena (formerly LMArena) is a community-powered platform created by UC Berkeley researchers for evaluating AI models through real-world human feedback. It features leaderboards for various AI capabilities including chat, text, image, video, vision, and more, allowing users to chat with models, compare them side-by-side, and contribute to rankings. Provides AI Evaluations service for enterprises and model labs. Supports use cases like chatting with AI, building apps, writing/editing text, web search, image/video generation and editing. Includes public leaderboard rankings for Overall, Agent, Text, WebDev, Image-to-WebDev, Text to Image, Image Edit, Text to Video, Image to Video, Video Edit, Vision, and Document Search.
How to get startedverified 8/25/2026
- Visit arena.ai and select models to chat with
- Provide feedback on model responses
- View and explore the public leaderboard rankings
What creators are saying
Sorted by extraction confidence, most recent first.
“I audited 100+ AI tools so you don’t have to. Here are the only 5 that matter. 🤖📊👇”I audited 100+ AI tools so you don’t have to. Here are the only 5 that matter. 🤖📊👇
Product details
Maintainer
UC Berkeley researchers
Links