• Initial context windowing with FreeNoise noise shuffling mainly for vid2vid and pose2vid pipelines for longer generations, haven't figured it out for img2vid yet
  • GGUF models and tiled encoding for I2V and pose pipelines (thanks to MinusZoneAI)
  • sageattention support (Linux only) for a speed boost, I experienced ~20-30% increase with it, stacks with fp8 fast mode, doesn't need compiling
  • Support CogVideoX-Fun 1.1 and it's pose models with additional control strenght and application step settings, this model's input does NOT have to be just dwpose skeletons, just about anything can work
  • Support LoRAs
    https://github.com/kijai/ComfyUI-CogVideoXWrapper/tree/main