AI · Web app
Diffusion Studio
Open-source video editor combining an infinite canvas with AI agent workflows
Diffusion Studio is an open-source video editor featuring an infinite spatial canvas, multitrack timeline, and built-in AI media generation. It connects with coding agents via MCP and CLI to let tools like Claude Code or Cursor analyze footage, cut clips, and generate audio or visuals directly in the project codebase.
What it's for
- Generate and replace visual elements in video scenes using AI image and video models
- Delegate video editing tasks like clipping, transcription, and sound design to coding agents
- Organize complex multimedia campaigns and multi-format scenes on an infinite canvas
- Export high-resolution 4K video compositions with keyframe easing and effects
Features
- Infinite canvas paired with a multitrack timeline editor
- Coding agent integration via Model Context Protocol (MCP) and CLI
- Built-in generation for images, video, sound effects, and voiceover
- Property inspector with bezier curve animation and layout control
- Support for HTML, WebGL, 3D elements, and shaders in compositions
- Non-linear multitrack video and audio editing with 4K export support
- Open-source codebase licensed under MPL-2.0
Pricing
Freemium. Core editor, CLI, and exports are free and open source; paid tiers purchase credits for hosted AI model generation
Details
- Type
- Web app
- Category
- AI
- Topic
- AI Video Editors
- Open source
- Yes
- Platforms
- Web, macOS, Windows, Command line
- For
- Video creators, motion designers, and developers building AI-assisted video pipelines
- Alternative to
- CapCut, Runway, Adobe Premiere Pro, Descript
- Tags
- video editor, ai video, infinite canvas, open source, agentic ai, mcp, motion design
Why it was saved
- @konstipaulus: Described as 'Figma meets Sora': a visual canvas where you drag assets, prompt an agent to generate/replace elements (e.g. swap a Minecraft block into an ASMR photo with FLUX.1 CONTEXT-MAXM, cut videos with KLING-2.1-Standard, add sound with MMAUDIO-V2), approve results, and edit on a Timeline with export.
Compare
Related
Open-source capability layer giving AI agents access to browse and scrape platforms
Open-source general AI agent framework replicating Manus without invite codes
Open-source browser-based AI video editor powered by 100+ models
AI-powered video editor that automates subtitles, cuts, and visual polish.
AI video editor for automated captions, B-roll, and short-form clips
AI-assisted multitrack video editor and text-to-video generator for creators and teams.
Updated