- Home
- LLM Evaluation Tools
- Arena (formerly Chatbot Arena)
Arena (formerly Chatbot Arena)
What it is
Arena is a free web platform where people compare AI model responses and vote to shape public leaderboards. In Battle mode, it displays two anonymous responses and reveals the model identities after a vote. Arena uses the Bradley-Terry rating system to combine votes into rankings; only anonymous Battle votes count toward official rankings. Side by Side and Direct modes are also available. Leaderboards cover agent, text, web development, image, video, vision, document, and search tasks, and the platform supports comparisons for image generation, image editing, and video generation, among other tasks. Leaderboards can be viewed without signing in, but Direct, Side by Side, Image Arena, and Video Arena require login. Arena applies a daily usage limit that resets 24 hours after it is reached. Video Arena requires login, allows two generations per 24-hour period, and offers Battle mode only. Arena says conversations may be used for research datasets and private model-provider evaluations, with shared data de-identified; it advises against submitting sensitive information. Commercial evaluation services are available to enterprises, model labs, and developers, with service-level agreements.
For and against
For
- Public leaderboards are viewable without signing in
- Battle votes shape official rankings
- Rankings cover multiple task categories
- Open research datasets are free to access
Against
- Daily usage limits apply to all users
- Video Arena requires login
- Video Arena allows two generations per 24 hours
- Conversations may be used for evaluations and datasets
The verdict
Arena offers public, vote-based comparisons across multiple model task areas, alongside commercial evaluation services. Keep the daily limits and potential use of conversations in mind, and avoid entering sensitive information.
Who it suits
Arena suits people who want to compare model responses or consult public leaderboards. Its commercial evaluation services are for enterprises, model labs, and developers.
Plans and prices
Public Arena
Public leaderboard and head-to-head battles are free; daily usage limit applies and resets after 24 hours
Read 7 Oct 2026 · arena.ai
AI evaluations
for enterprises, model labs and developers · grounded in real-world human feedback
Read 21 Oct 2026 · arena.ai
On the record
- Account benefits
- Creating an account unlocks Direct, Side-by-Side, Image, and Video modes, higher rate limits, and synchronized chat history.help.arena.ai · read 21 Oct 2026
- Account integrations
- Users can create an account with Google or email, and Arena’s coding Agent Mode can connect to GitHub repositories.help.arena.ai · read 7 Oct 2026
- Account requirements
- Leaderboards can be viewed without signing in, while Direct, Side by Side, Image Arena, and Video Arena require login.help.arena.ai · read 7 Oct 2026
- Agent files
- Agent Mode accepts PNG, WebP, JPEG, PDF, GIF, TXT, Markdown, CSV, HTML, XML, CSS, JavaScript, and JSON uploads.help.arena.ai · read 21 Oct 2026
- Agent mode
- Agent Mode plans multi-step work and can use web search, image generation, file uploads, coding assistance, and a sandbox/bash environment.help.arena.ai · read 21 Oct 2026
- Age restriction
- The privacy policy says the service is not intended for people under 18.help.arena.ai · read 21 Oct 2026
- Anonymous battles
- In Battle mode, Arena presents two anonymous model responses and reveals their identities after the user votes.arena.ai · read 7 Oct 2026
- API
- Arena provides an OpenAI-compatible gateway with model listing and chat-completion endpoints authenticated by virtual API keys.portal.api.preview.arena.ai · read 21 Oct 2026
- API compatibility
- The API documentation shows Anthropic-compatible client configuration for Claude Code through the Arena API base URL.portal.api.preview.arena.ai · read 21 Oct 2026
- Battle mode
- Users submit a prompt, receive two anonymous model answers, vote, and then see the model identities.arena.ai · read 21 Oct 2026
- Commercial customers
- Arena offers commercial evaluation services to enterprises, model labs, and developers, with evaluation results supported by service-level agreements.arena.ai · read 7 Oct 2026
- Company origin
- Arena says it was created by UC Berkeley researchers and that Arena Intelligence Inc. was formed in 2025.arena.ai · read 7 Oct 2026
- Data sharing
- Arena says conversations may be de-identified and shared in public research datasets or private evaluations with model providers.arena.ai · read 21 Oct 2026
- Developer documentation
- The public API reference identifies itself as OpenAPI 3.0.3 and includes curl quickstarts.portal.api.preview.arena.ai · read 21 Oct 2026
- Discord
- The terms cover an Arena Discord server and an Arena Discord Bot as part of the service.help.arena.ai · read 21 Oct 2026
- Funding
- Arena says it has $250 million in venture funding led by Felicis and UC Investments, with participation from several other investors.arena.ai · read 21 Oct 2026
- GitHub integration
- When a GitHub repository is connected, Agent Mode works in a copy, commits changes to a branch, and opens a pull request.help.arena.ai · read 21 Oct 2026
- Independent research result
- The Chatbot Arena paper reports more than 240,000 votes and says crowdsourced questions were diverse and human votes agreed well with expert ratings.arxiv.org · read 21 Oct 2026
