Video generation and diffusion models

8 repos

Libraries, model implementations, and optimizations for text-to-video and video generation using diffusion models. The cluster centers on the Wan2.1 and VACE model families alongside supporting frameworks like diffusers, with tooling for model serialization (safetensors) and inference across different model scales. Most repositories focus on Chinese and English language support, addressing both research implementations and production-ready model variants.

Related clusters