ComfyUI Extension: ComfyNodesPlus
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 growing collection of practical, high-quality nodes designed to extend ComfyUI with useful utilities, smarter workflow controls, and flexible generation tools.
Looking for a different extension?
Custom Nodes (0)
README
ComfyNodesPlus
Enhanced custom nodes for ComfyUI.
ComfyNodesPlus is a growing collection of practical, high-quality nodes designed to extend ComfyUI with useful utilities, smarter workflow controls, and flexible generation tools. Built with a focus on simplicity, power, and real-world usability.
β¨ Features
- π² Smart randomisation nodes (resolutions, ratios, seeds)
- βοΈ Workflow utility and control nodes
- π§© Modular and easy to extend
- π Lightweight and performance-friendly
- π§ Designed for both beginners and advanced pipelines
π¦ Installation
Clone into your ComfyUI custom_nodes directory:
cd ComfyUI/custom_nodes
git clone https://github.com/chemicalshock/ComfyNodesPlus.git
Restart ComfyUI after installation.
π Included Nodes
π² Random Image Size
Generate dynamic image resolutions for more varied outputs.
Inputs:
min_sizeβ Minimum dimension (e.g. 512)max_sizeβ Maximum dimension (e.g. 2048)modeβfixedorrandomorientationβsquare,vertical,horizontal, orrandom
Outputs:
widthheight
Example outputs:
- 512 Γ 512
- 512 Γ 1024
- 1024 Γ 1024
- 2048 Γ 1024
(More nodes coming soon.)
π§± Project Structure
ComfyNodesPlus/
βββ nodes/
β βββ CNP_random_image_size.py
β βββ ...
βββ __init__.py
βββ requirements.txt
βββ README.md
π§© Creating New Nodes
All nodes follow standard ComfyUI conventions:
class ExampleNode:
@classmethod
def INPUT_TYPES(cls):
return {
"required": {
"value": ("INT", {"default": 0})
}
}
RETURN_TYPES = ("INT",)
FUNCTION = "execute"
def execute(self, value):
return (value,)
Register nodes in __init__.py.
π― Design Principles
- Keep nodes focused and useful
- Avoid unnecessary complexity
- Prioritise real workflow value
- Support both deterministic and randomised pipelines
- Maintain clean, readable code
π§ Requirements
- ComfyUI (latest recommended)
- Python 3.10+
π€ Contributing
Contributions are welcome.
Please:
- Keep code clean and minimal
- Follow the existing structure
- Document nodes clearly
- Avoid unnecessary dependencies
π License
MIT
π§ Vision
ComfyNodesPlus aims to provide a reliable, growing toolkit of nodes that make ComfyUI workflows faster, smarter, and more expressive.
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.