This repository provides a collection of nodes designed for ComfyUI that automate the adjustment of sampler settings based on model names and facilitate image resizing with specific constraints. Its primary aim is to enhance workflow efficiency and control when working with image generation and processing.
- NodeAutoSampler: Automatically configures sampler parameters based on the model's checkpoint name using conditional logic.
- NodeImageResize: Resizes images while preserving their aspect ratio, ensuring dimensions are rounded up to the nearest multiple of 16.
- NodeImagePre: Automatically generates resizing parameters, empty latent space, and upscaling factors based on the target short side of an image.
Context
This tool is a specialized node package for ComfyUI that streamlines the process of adjusting image generation parameters and resizing images. It allows users to automate settings based on model names and ensures that images conform to specific size requirements, thereby simplifying the workflow for artists and developers working with AI-generated content.
Key Features & Benefits
The primary features include the ability to automatically set sampler configurations based on model names, which saves time and reduces manual errors. Additionally, the image resizing nodes ensure that images maintain their aspect ratios while adhering to specific size constraints, which is crucial for compatibility with various models and outputs.
Advanced Functionalities
The NodeAutoSampler utilizes conditional logic to determine sampler settings based on keywords found in the model's checkpoint name. This allows for dynamic adjustments, such as lowering steps or CFG values when certain keywords are present. The NodeImagePre node further enhances usability by intelligently calculating necessary parameters for resizing and upscaling, thereby optimizing the image preparation process.
Practical Benefits
This tool significantly improves the efficiency of workflows in ComfyUI by automating repetitive tasks related to sampler settings and image resizing. Users gain greater control over their outputs, ensuring that images are not only correctly sized but also optimized for the specific models being employed. This leads to higher quality results and a more streamlined creative process.
Credits/Acknowledgments
The tool is developed by KERRY-YUAN and is available under an open-source license, allowing for community contributions and updates.