ComfyUI Extension: ComfyUI-HunyuanVideoImagesGuider
Run ComfyUI workflows without the setup
No installs, no CUDA version roulette, no GPU sitting idle on your bill. Bring a workflow and run it in the browser.
A specialized node for ComfyUI that enable advanced motion and animation capabilities for image as guider for video processing In Hunyuan Video.
README
ComfyUI-HunyuanVideoImagesGuider
A specialized node for ComfyUI that enable motion and animation capabilities for image as guider for video processing In Hunyuan Video T2V model weights.
How to use this node , you can check it out in this video : https://youtu.be/fq95OLDVCdU
Features
Hunyuan Video Image To Guider
- Smooth camera panning effects (X and Y axis)
- Zoom in/out animations
- Seamless image tiling and repetition
- Customizable frame count and motion strength
- Support for various resize modes
- Optional center cropping
Installation
- Clone this repository into your ComfyUI custom nodes directory:
cd ComfyUI/custom_nodes
git clone https://github.com/benjiyaya/ComfyUI-HunyuanVideoImagesGuider.git
- Restart ComfyUI to load the new nodes
Usage
Hunyuan Video Image To Guider
Parameters:
image: Input imagemove_range_x: Horizontal motion range (-1.0 to 1.0) (Best range in -0.05 - 0.05)move_range_y: Vertical motion range (-1.0 to 1.0) (Best range in -0.05 - 0.05)zoom: Zoom effect strength (0.0 to 0.5)frame_num: Number of frames to generate (2-150)resize_mode: Image resize options (disabled/custom/keep_ratio)target_width: Custom width for resizetarget_height: Custom height for resizecenter_crop: Enable/disable center cropping
*** For the sampler denoise, good range in 0.7-0.9 , below 0.5 almost no animation for object.
Technical Details
The nodes utilize PyTorch for efficient tensor operations and implement various optimization techniques:
- Automatic mixed precision
- Efficient memory management
- Batched processing
- Error handling and recovery
- Progress tracking

https://github.com/user-attachments/assets/7701fdec-74db-4277-a883-3c204037e5ba

https://github.com/user-attachments/assets/e9bec99d-f87e-44b2-95c2-48113085e39b

https://github.com/user-attachments/assets/0b88a0d3-9071-4cb8-adae-d47776f27cdd

https://github.com/user-attachments/assets/e5a8d3bb-f43b-4a87-bdde-926afd03cd1b
https://github.com/user-attachments/assets/77bf3cbc-8292-466b-9fff-ae4b7ec2c835
Contributing
Contributions are welcome! Please feel free to submit a Pull Request.
License
MIT License - feel free to use in your own projects.
Credits
Created by Benji and DeepSeek AI Copilot
Run ComfyUI workflows without the setup
No installs, no CUDA version roulette, no GPU sitting idle on your bill. Bring a workflow and run it in the browser.