ali-vilab/TeaCache
Speeds up video and image diffusion models by caching computations, no retraining needed
View on index · View in 3D Map
Related repos
- video diffusion model for text-to-video, image-to-video, extension, and transitions
- Real-time diffusion-based video super-resolution
- open-source diffusion engine for AI image and video generation
- diffusion transformer models for video restoration
- Training large-scale diffusion models on a budget
- open source image resizing and caching API for your own server
- One-step video restoration with a diffusion transformer
- video and image watermarking code for training and inference
- Hybrid cache with resiliency features
- Optimizes images in the browser without data leaving the device
