SeargeSDXL

835

Documentation

Searge-SDXL: EVOLVED v4.x for ComfyUI

Overview

The Searge-SDXL: EVOLVED v4.x is a custom nodes extension tailored for ComfyUI, specifically designed to enhance workflows that utilize SDXL 1.0 with both the base and refiner checkpoints. This extension offers a comprehensive set of features that streamline the workflow process and enable more advanced image generation and manipulation capabilities.

1. How to Install

New and Recommended Installation (Windows)

  1. Prerequisites: Ensure you are using the portable version of ComfyUI for Windows.
  2. Download Installer: Obtain the installation script from SeargeSDXL Installer.
  3. Unpack the Script: Extract the contents into the ComfyUI_windows_portable directory. The files SeargeSDXL-Installer.bat and SeargeSDXL-Installer.py should be placed alongside run_cpu.bat and run_nvidia_gpu.bat.
  4. Verify: Ensure the presence of the python_embeded directory in the same location.
  5. Run Installer: Execute SeargeSDXL-Installer.bat and follow the on-screen instructions.

Manual Installation

  1. Install Dependency: In the ComfyUI Python environment, run python -m pip install opencv-python.
  2. Clone Repository:
    • Navigate to the ComfyUI/custom_nodes/ directory.
    • Open a command line window in this directory.
    • Execute git clone https://github.com/SeargeDP/SeargeSDXL.git.
    • Restart ComfyUI.

Alternative Installation (Not Recommended)

  1. Download and Extract: Obtain the latest release from the Searge SDXL CivitAI page.
  2. Copy Files: Place the SeargeSDXL folder into the ComfyUI/custom_nodes directory.
  3. Restart ComfyUI.

Updating an Existing Installation

  • If using git, navigate to ComfyUI/custom_nodes/, open a command line, and run git pull.
  • If using a zip file, download the latest release, extract, and replace files in ComfyUI/custom_nodes.
  • Restart ComfyUI.

2. What This Repository is For

This repository provides a custom nodes extension for ComfyUI that facilitates the use of SDXL 1.0 with enhanced workflows leveraging both the base and refiner checkpoints. It is designed to optimize image processing pipelines, enable controlnett and revision functionalities, and improve image quality through various advanced features.

3. What Kinds of Nodes It Provides

The repository includes nodes for various functionalities:

  • Controlnet Nodes: Enable detailed control over image generation through different model checkpoints and annotators.
  • LoRA Nodes: Provide multi-LoRA support for layered image manipulation.
  • Scaler Nodes: Integrate upscalers like ESRGAN for enhanced image resolution.
  • Inpainting Nodes: Facilitate selective image modification using masks.
  • Debugging Nodes: Aid in pipeline debugging like the debug printer node.

4. Special Features or Capabilities

  • Integrated Workflow: A single comprehensive workflow file integrates all features, eliminating the need for separate workflows.
  • High Compatibility: Supports FreeU v2 alongside FreeU v1, providing options for different generation presets.
  • Advanced Processing Pipeline: Offers a highly optimized architecture, resulting in up to 20% faster image processing.
  • Control and Revision Support: Allows application of up to 5 controlnets and revisions simultaneously.
  • Multi-LoRA and High-Res Modes: Capabilities that enhance image generation and make it possible to use multiple LoRAs and improved high-resolution outputs.
  • Capability for Text, Image, and Inpainting Processing: Supports various modes including text-to-image, image-to-image, and inpainting operations.

5. How it Might Be Useful in ComfyUI Workflows

The Searge-SDXL repository enriches ComfyUI workflows by:

  • Streamlining Workflows: Integrates multiple functionalities into a single workflow, simplifying the setup and use process.
  • Enhancing Image Quality: Utilizes advanced nodes and processing techniques to improve the quality and detail of generated images.
  • Flexibility: Provides users with the ability to switch between various modes and presets, catering to different image generation needs.
  • Time-Efficiency: The optimized processing pipeline reduces rendering times, making it suitable for workflows requiring rapid iterations.

The Searge-SDXL extension for ComfyUI is ideal for users seeking a robust and versatile image generation setup, enhancing both the capability and efficiency of workflows constructed within ComfyUI.