Hunyuan Video Framepack

Kijai has released a comfyui custom node wrapper which allows us to start using Framepack.
Framepack is able to create sequences of latents in a batch, efficiently. That means longer videos.

YOU WILL NEED:

The standard Hunyuan Video CLIP, VAE and VISION models

The FramePack i2v model
FP8: https://huggingface.co/Kijai/HunyuanVideo_comfy/blob/main/FramePackI2V_HY_fp8_e4m3fn.safetensors
BF16: https://huggingface.co/Kijai/HunyuanVideo_comfy/blob/main/FramePackI2V_HY_bf16.safetensors

WORKFLOWS

https://github.com/MushroomFleet/DJZ-Workflows/tree/main/Hoda-Hunyuan-Video/FramePack