Newsletter
Join the Community
Subscribe to our newsletter for the latest news and updates
A next-frame prediction neural network for practical video generation with constant workload regardless of video length.
Discover the easiest way to create engaging videos online, and add audio visualizers, subtitles, effects, filters, and elements. In your browser, no app required! Try it now, Unlimited free videos.
Turn your music into a killer video. Specterr makes it fast and easy to create custom music visualizers. Start growing your audience with video.
FramePack is an innovative neural network structure designed for next-frame (or next-frame-section) prediction in video generation. It compresses input contexts to a constant length, making the generation workload invariant to video length. This allows processing a large number of frames even on laptop GPUs with 13B models. FramePack can be trained with batch sizes similar to image diffusion training, making video diffusion as practical as image diffusion.
Key features:
Use cases: