OpenChatCut0.2.17

OpenChatCut lets you describe an edit conversationally and have the resulting tracks, clips, transitions, captions, effects, and media written directly into the video project. It combines a conversational editing agent with a professional multi-track timeline, so you can inspect and adjust the agent’s work instead of treating its response as a separate set of instructions. The local-first design keeps projects and media on your machine by default while preserving direct control over each project element.

The built-in agent and external MCP agents use the same editing tools, allowing the project to serve as shared working space for conversational and hands-on editing. You can make manual changes, undo or redo operations, save project versions, and hand the work to another agent. The timeline supports transitions, effects, LUTs, zooms, keyframes, Agent Skills, and Remotion-based rendering.

Word-level transcription enables transcript-driven editing and text-based cuts from spoken content. OpenChatCut can also generate images, video, speech, and music, while WebGL motion graphics, in-browser person segmentation, and face-safe zones provide additional visual controls. Export formats include MP4, audio, captions, FCPXML, and complete project data.

Published by 0xsline for Windows, version 0.2.17 runs on x64 systems with Windows 10 or later or Windows Server 2016 or later, and is open-source. This release fixes native ASR model resolution through the shared catalog and adds OrcaRouter as a named LLM provider preset.