Install this extension via the ComfyUI Manager by searching
for m957ymj75urz/ComfyUI-Custom-Nodes
1. Click the Manager button in the main menu
2. Select Custom Nodes Manager button
3. Enter m957ymj75urz/ComfyUI-Custom-Nodes in the search bar
After installation, click the Restart button to
restart ComfyUI. Then, manually
refresh your browser to clear the cache and access
the updated list of nodes.
Visit
ComfyUI Online
for ready-to-use ComfyUI environment
m957ymj75urz/ComfyUI-Custom-Nodes offers specialized nodes like RawText, RawTextCLIPEncode, RawTextCombine, and RawTextReplace, along with the m957ymj75urz.colors extension, enhancing text processing and customization in ComfyUI.
m957ymj75urz/ComfyUI-Custom-Nodes Introduction
ComfyUI-Custom-Nodes is an extension designed to enhance the functionality of ComfyUI, a powerful and modular interface for creating and executing advanced Stable Diffusion workflows. This extension introduces a variety of custom nodes that can be used to manipulate images, text, and other data types within the ComfyUI environment. By using these custom nodes, AI artists can streamline their creative processes, experiment with new techniques, and achieve more refined results without needing to write code.
How m957ymj75urz/ComfyUI-Custom-Nodes Works
ComfyUI-Custom-Nodes operates by adding new nodes to the ComfyUI interface. These nodes can be thought of as building blocks that you can connect together to create complex workflows. Each node performs a specific function, such as loading images, applying aesthetic filters, or generating text prompts. By connecting these nodes in different ways, you can create a wide range of effects and automate various parts of your creative process.
Imagine you are building a pipeline where each step is a node. For example, you might start with a node that loads an image, then connect it to a node that applies a filter, and finally connect it to a node that saves the modified image. This modular approach allows you to easily experiment with different configurations and see the results in real-time.
m957ymj75urz/ComfyUI-Custom-Nodes Features
Aesthetic Nodes
Aesthetic Loader: Loads models for use with the Aesthetic Select node.
Aesthetic Select: Returns the best tensors based on aesthetic, style, waifu, or age classifiers. If no models are selected, it acts like a basic node and returns a single tensor. Setting the count to 0 stops processing for connected nodes.
IF Nodes
IF Load: Loads models for use with other IF nodes. You can specify the device (e.g., CPU, CUDA) to move the models to.
IF Encode: Encodes prompts for use with IF Stage I and IF Stage II nodes.
IF Stage I: Takes the prompt embeds from IF Encode and returns images, which can be used with IF Stage II or other nodes.
IF Stage II: Takes images from IF Stage I and upscales them by 4x.
IF Stage III: Upscales images using the Stable Diffusion x4 upscaler. This node has its own encoder and accepts string prompts.
Image Nodes
Image Batch: Loads all images in a specified directory as a batch. The images will be cropped or resized if their dimensions aren't equal.
Image Saver: Saves images without metadata in a specified directory. Allows saving a batch of images as a grid or animated GIF.
Multi Nodes
Multi Crop: Center crops or pads tensors to specified dimensions.
Multi Noise: Adds random noise to tensors.
Multi Repeat: Repeats tensors a specified number of times.
Multi Resize: Resizes tensors using a scale factor instead of width/height.
Text Nodes
Text Loader: Initializes the Text Prompt node so you don't have to specify the same parameters multiple times.
Text Prompt: Queries an API with parameters from Text Loader and returns a string for use with other nodes.
Text Condition: Returns input tensors and a boolean value based on whether variables match a condition.
Text Format: Joins input strings with multiple variables and returns a single output string.
Text Print: Prints string input to the console for debugging purposes.
m957ymj75urz/ComfyUI-Custom-Nodes Models
Aesthetic Models
Aesthetic: Classifies images based on aesthetic quality.
Style: Classifies images based on style.
Waifu: Classifies images based on anime-style characters.
Age: Classifies images based on the apparent age of subjects.
IF Models
DeepFloyd IF: A set of models for generating and upscaling images based on text prompts. These models are automatically downloaded but require agreeing to terms of use and logging in with an access token.
What's New with m957ymj75urz/ComfyUI-Custom-Nodes
Updates and Changes
Archival Notice: The repository is now archived and will no longer be maintained or updated. Most nodes have been integrated into ComfyUI or exist elsewhere.
Aesthetic Nodes: Will be moved to a separate repository in the future.
Troubleshooting m957ymj75urz/ComfyUI-Custom-Nodes
Common Issues and Solutions
Installation Issues: If requirements are not installed automatically, run pip install -r custom_nodes\Zuellni\requirements.txt.
IF Nodes Not Loading: Ensure you have agreed to the terms of use, created an access token, and logged in.
Image Batch Loading Errors: Make sure all images in the directory have equal dimensions or enable cropping/resizing.
Frequently Asked Questions
How do I update the extension?
Set Update Repository to true in the config file or run git -C custom_nodes\Zuellni pull.
Can I disable certain nodes?
Yes, set IF to false under Load Nodes in the config file to disable IF nodes.
Learn More about m957ymj75urz/ComfyUI-Custom-Nodes
Additional Resources
ComfyUI Examples: Explore various workflow examples to see what ComfyUI can do.
Community Forums: Join discussions, ask questions, and get support from other users.
By leveraging the power of ComfyUI-Custom-Nodes, you can unlock new creative possibilities and streamline your workflow, making it easier to produce high-quality AI-generated art.