Various custom nodes designed to enhance the functionality of ComfyUI, focusing on resolution selection, random seed generation, and conditional LoRA application based on prompts.
- Provides a manual and random resolution selection feature, allowing users to customize their output dimensions easily.
- Includes a Conditional LoRA Loader that applies specific LoRAs based on regex patterns from external configuration files, enabling dynamic content adaptation.
- Generates lists of random seed values, facilitating varied and unpredictable outputs in image generation tasks.
Context
This repository contains a collection of custom nodes for ComfyUI, aimed at providing users with enhanced control over image resolution selections and the dynamic application of LoRAs (Low-Rank Adaptation models). These nodes are particularly useful for artists and developers looking to streamline their workflows in AI-generated art.
Key Features & Benefits
The Resolution Selector node allows users to manually choose image dimensions from a customizable dropdown menu, which can be populated from a user-defined JSON file. The Random Resolution Selector node dynamically selects resolutions based on a seed input, ensuring that users can generate varied outputs without manually changing settings each time. The Conditional LoRA Loader applies specific LoRAs based on regex matches against incoming prompts, allowing for targeted enhancements to generated images.
Advanced Functionalities
The Conditional LoRA Loader is particularly advanced, as it can evaluate multiple regex patterns from an external TOML configuration file to determine which LoRAs to apply. This means that users can create complex rules for when certain LoRAs should be activated, providing a high level of customization and control over the output based on the content of the prompts. Additionally, the SeedListGenerator simplifies the process of creating random seed lists, which can be crucial for generating diverse outputs in artistic workflows.
Practical Benefits
These custom nodes significantly enhance the workflow within ComfyUI by providing greater flexibility in resolution selection and enabling dynamic content adaptation through the Conditional LoRA Loader. Users can expect improved efficiency and quality in their outputs, as they can easily manipulate resolution settings and apply relevant LoRAs based on specific prompt conditions, ultimately leading to a more tailored and high-quality image generation process.
Credits/Acknowledgments
This project is licensed under the MIT License, and the original authors and contributors are acknowledged within the repository. Users are encouraged to refer to the LICENSE file for more details on usage and contributions.