LaVie
Text-to-video generation framework with cascaded latent diffusion.
About
LaVie is a research text-to-video framework from Shanghai AI Lab and Vchitect that uses cascaded latent diffusion models with temporal super-resolution to produce short video clips at higher resolution than the base generator alone. The repository is the official PyTorch implementation accompanying the paper and ships pretrained model weights, an image-to-video companion (SEINE), and a Hugging Face demo space.
Reviews (0)
Leave a Review
No reviews yet. Be the first to review!
Details
- Category
- Video Generation
- Price
- Free
- Platform
- Local/Desktop
- Difficulty
- Advanced (4/5)
- License
- Apache-2.0
- Minimum VRAM
- 16 GB
- Added
- Apr 3, 2026
Related Tools
Latest video generation model by Alibaba with improved quality and speed.
Infinite-length music-driven video generation with visual conditioning.
Image-to-video generation model by Alibaba DAMO Academy.
Open-source video generation model with controllable camera and subject motion.
Updated video generation model by Tencent with improved quality.
Updated CogVideo model by Zhipu AI with improved video quality.