(Tutorial)Hunyun Video Img2vid In ComfyUI - Support In Day 1 Native Node
Added 2025-03-07 13:00:12 +0000 UTC
Video : https://youtu.be/XsEi02yH_hI
Discover the newly released Hanyuan Image-to-Video model, now supported in ComfyUI! This tutorial walks you through setting up the workflow, optimizing resolutions, and using dual sampling for smoother, higher-quality outputs. Perfect for creating dynamic character animations, landscapes, and more—all on your local machine!
Who is this content suitable for?
AI developers, video creators, and ComfyUI users looking to explore advanced image-to-video workflows. Ideal for those working with Hanyuan Video models or seeking to enhance video quality with dual sampling techniques.
Why does it matter?
The Hanyuan Image-to-Video model brings cinematic quality to local AI workflows, offering smoother motions, natural character movements, and realistic landscapes. Whether creating character animations or scenic videos, this model empowers creators to produce professional-grade content without relying on cloud services.
Hunyuan Image2Video in ComfyUI:
https://docs.comfy.org/advanced/hunyuan-video
Hugging Face:
https://huggingface.co/Comfy-Org/HunyuanVideo_repackaged/tree/main/split_files
Download the following models and place them in the locations specified below:
Copy in ComfyUI/Models/
├── clip_vision/ │
└── llava_llama3_vision.safetensors
├── text_encoders/ │
├── clip_l.safetensors │
├── llava_llama3_fp16.safetensors │
└── llava_llama3_fp8_scaled.safetensors
├── vae/ │
└── hunyuan_video_vae_bf16.safetensors
└── diffusion_models/
└── hunyuan_video_image_to_video_720p_bf16.safetensors
Attached Img2Vid basic workflow, have fun :)
