ComfyUI diffusers wrapper nodes for [a/HunyuanVideo](https://github.com/Tencent/HunyuanVideo)

ComfyUI-HunyuanVideoWrapper

Loads Hunyuan text_encoder model from 'ComfyUI/models/LLM'

DownloadAndLoadHyVideoTextEncoder

Settings for block swapping, reduces VRAM use by swapping blocks to CPU memory

HyVideoBlockSwap

HyVideoCFG

Context options for HunyuanVideo, allows splitting the video into context windows and attemps blending them for longer generations than the model and memory otherwise would allow.

HyVideoContextOptions

HyVideoCustomPromptTemplate

HyVideoDecode

Empty Text Embeds for HunyuanVideoWrapper, to avoid having to encode prompts for inverse sampling

HyVideoEmptyTextEmbeds

HyVideoEncode

HyVideoEncodeKeyframes

https://github.com/NUS-HPC-AI-Lab/Enhance-A-Video

HyVideoEnhanceAVideo

HyVideoGetClosestBucketSize

HyVideoI2VEncode

HyVideoInverseSampler

HyVideoLatentPreview

HyVideoLoraBlockEdit

Select a LoRA model from ComfyUI/models/loras

HyVideoLoraSelect

HyVideoModelLoader

HyVideoPromptMixSampler

HyVideoReSampler

HyVideoSampler

Spatio Temporal Guidance, https://github.com/junhahyung/STGuidance

HyVideoSTG

TeaCache settings for HunyuanVideo to speed up inference

HyVideoTeaCache

Acts as a bridge between the native ComfyUI conditioning and the HunyuanVideoWrapper embeds

HyVideoTextEmbedBridge

HyVideoTextEmbedsLoad

HyVideoTextEmbedsSave

HyVideoTextEncode

HyVideoTextImageEncode

torch.compile settings, when connected to the model loader, torch.compile of the selected layers is attempted. Requires Triton and torch 2.5.0 is recommended

HyVideoTorchCompileSettings

Loads Hunyuan VAE model from 'ComfyUI/models/vae'

ComfyUI Extension: ComfyUI-HunyuanVideoWrapper

Custom Nodes (29)