1 repository
Techniques and toolkits for accelerating diffusion model inference through caching, quantization, and memory management.
Distinct from Diffusion Models: Distinct from Diffusion Models: focuses on the optimization and acceleration layer for existing models rather than the model architectures themselves.
Explore 1 awesome GitHub repository matching artificial intelligence & ml · Optimization Frameworks. Refine with filters or upvote what's useful.
FramePack is a neural video synthesis engine and generation framework designed to produce long, temporally consistent video sequences. It functions as a diffusion model optimizer, providing a suite of techniques to manage the computational demands of high-parameter video models while maintaining visual stability during extended generation tasks. The system distinguishes itself through a hierarchical approach to frame prediction, which plans distant anchor frames before filling in intermediate content to prevent cumulative temporal drift. By utilizing constant-length context compression and to
Provides a suite of optimization techniques including caching and quantization to accelerate diffusion-based video generation on consumer hardware.