v1.52: Live Voice Transcription
Voice input now streams live as you speak, and you can attach text files directly to conversations.
Voice input now streams live as you speak, and you can attach text files directly to conversations.
Fabric can now run multiple AI agents in parallel and coordinate complex workflows automatically. Plus improved command search and better browser integration.
Fabric can now control your browser—opening tabs, clicking buttons, filling forms, and taking screenshots—all driven by AI.
Find any setting instantly with the new searchable settings panel, plus improved chat history navigation and better tab management.
Type @ in chat to search and attach project files as context. Fabric now remembers your window position and split-pane layout across sessions.
Work on multiple conversations side-by-side with split-pane tiling. Divide your workspace into up to 8 independent panes, each with its own chat session and tabs.
Fabric now automatically hides AI thinking and intermediate work, showing only the final answer. Plus, Take the Wheel mode is faster and more reliable.
Fabric now fully supports IME input for Japanese, Chinese, and Korean languages, plus improved permission controls and media previews.
Attach video files to your messages, install MCP servers directly from chat, and reorder tabs with drag-and-drop.