The page documents notable changes to the arena.ai product including updates, new features, and more.
August 17, 2026
- Agent Arena leaderboard improvements: Measuring task cost, new agent Pareto view, and new categories in Agent Arena.
July 30, 2026
- AutoEvals are now live, providing a new evaluation methodology that ranks models using reward models based on millions of real Arena user preferences.
July 14, 2026
- Factuality: ranking models not only by human preference, but also by the factual accuracy of their responses.
July 2, 2026
- Fullstack Code Arena: databases, authentication, third-party integrations, and deployments, to build and deploy production-ready applications.
June 4, 2026
- Agent Mode: Empowering users to get more done with agentic capabilities.
May 8, 2026
- Introduced 7 new categories for web development in Code Arena.
May 5, 2026
- Multimodal Max: Expanded capabilities including search, vision, image generation, image editing, and front-end coding.