Summary: LMArena adds multimodal leaderboards and new models

Why It Matters: The platform has expanded from simple text comparisons to ranking how AI models handle images, video, search, and coding. It also reveals the names of several unreleased or next-generation models.

What Changed:

• Added new leaderboard categories for Vision, Text-to-Image, Image Edit, Search, Text-to-Video, and Image-to-Video.
• Updated the main rankings with new high-performing models like Gemini 3 Pro, Grok 4.1, and Claude 4.5.
• Introduced a "Leaderboard Overview" page that provides a snapshot of model performance across all different arenas.
• Added a detailed "Arena Overview" table that ranks models by specific skills like Math, Creative Writing, and Long Queries.