Repolode
Submit
glanderness

glanderness/BeefTV

A desktop AI-native video creation workspace where you connect ideas, models and assets on a free-form canvas, with an agent taking part in the workflow.

Open-source alternative to:RunwayAdobe FireflyCanva

View on GitHubWebsite
Stars
1,511
Forks
291
Open issues
24
Last commit
yesterday

What is it?

BeefTV is a video creation workspace built around an infinite canvas. You place prompts, reference materials and generated results as nodes and connect them with lines to show how they relate, so a generation flow stays visible and editable instead of living in a chat log.

On the canvas you can generate text, images, video and audio from prompts or reference assets, then crop, annotate, partially redraw, split and combine the results. Projects, a personal asset library, asynchronous generation tasks and model channels are managed in one workspace, and data can be stored locally and migrated.

The model side is configurable: text, image, video and audio model channels can be added without being tied to a single provider. The desktop app is built with React, Go and Wails for macOS and Windows.

Key features

  • 01Infinite canvas where prompts, assets and generated results are connected as nodes.
  • 02Generates text, image, video and audio from prompts or reference materials.
  • 03Crop, annotate, partial redraw, split and combine generated results.
  • 04Configurable model channels that are not locked to one provider.
  • 05Project library, personal asset library and asynchronous generation tasks in one workspace.
  • 06Local data storage with migration support.

Use cases

  • Building a short video or storyboard by chaining generated images and clips on one canvas.
  • Reusing reference assets and iterating on generated results without losing the creation history.
  • Experimenting with different image, video or audio models in the same workflow.

Topics

#ai-video#generative-ai#creative-tools#infinite-canvas#local-first#aigc#ai-image-generation#ai-video-generation#ai-workspace

FAQ

What is BeefTV?
A desktop AI-native video creation workspace where you connect ideas, models and assets on a free-form canvas, with an agent taking part in the workflow. BeefTV is a video creation workspace built around an infinite canvas. You place prompts, reference materials and generated results as nodes and connect them with lines to show how they relate, so a generation flow stays visible and editable instead of living in a chat log.
Is BeefTV free to use? What license does it have?
BeefTV is open source under the MIT license. Check the license terms before using it in commercial products.
Is BeefTV still maintained?
Actively maintained. The last commit to BeefTV was on Oct 17, 2026.
What can I use BeefTV for?
Building a short video or storyboard by chaining generated images and clips on one canvas. Reusing reference assets and iterating on generated results without losing the creation history. Experimenting with different image, video or audio models in the same workflow.
Where can I get BeefTV?
The source code and installation instructions are on GitHub: https://github.com/glanderness/BeefTV

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects