From idea to finished cut
Every tool, AI workflow and export option in ChatCut: where to find it and what it does to your timeline. Chapters follow the order you'll need them.
Chapters
Understand ChatCut What ChatCut does, which version to use, and your first edit. 
AI Workspace Prompts, Selection Mode, Skills, Styles, voices and approvals. 
Editor and Media Assets, Library, timeline, transcript, captions, recording and versions. 
AI Features Video, images, motion graphics, voice, music, avatars and translation. 
Export and Delivery Video, captions, transparent graphics and editable project files. 
Desktop app Local media, desktop agents, local export and 4K. 
Agent Plugin Connect ChatCut to Claude Code or Codex. 
Account, Collaboration, and Billing Sign-in, sharing, plans, credits and usage. 
Troubleshooting and Reference Errors, permissions, formats, model limits and shortcuts. Create a project Open ChatCut and start editing Changelog
View all releases- Desktop projects, URL imports, and direct effect editing Import cloud projects and online media with visible progress, edit effect geometry in the preview, and keep local projects more reliable.
- Real caption previews, project-media choices, and script fixes Compare caption styles with project text, choose project media visually in Agent forms, and avoid overlapping or fragmented transcript edits.
- Caption style previews, zoom presets, and localized effects See caption styles before applying them, use clearer talking-head zoom presets, and find effects with localized names and controls.
- GPT Image 2.5, spoken corrections, and Linear Mask Choose new image models and quality options, repair short spoken mistakes with a matching voice, and shape effects with a native linear mask.
- Motion-graphic element editing and a unified voice library Select and restyle individual motion-graphic elements in Desktop, and choose from a clearer shared voice catalog across generation workflows.
- MiniMax H3 video, vocal songs, translation review, and Hermes Agent Generate with two new video models, create vocal songs, review source transcripts before lip sync, and run Hermes inside Desktop.