ComfyUI  >  Nodes  >  PuLID_ComfyUI

ComfyUI Extension: PuLID_ComfyUI

Repo Name


cubiq (Account age: 5009 days)
View all nodes (5)
Latest Updated
Github Stars

How to Install PuLID_ComfyUI

Install this extension via the ComfyUI Manager by searching for  PuLID_ComfyUI
  • 1. Click the Manager button in the main menu
  • 2. Select Custom Nodes Manager button
  • 3. Enter PuLID_ComfyUI 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

  • Free trial available
  • High-speed GPU machines
  • 200+ preloaded models/nodes
  • Freedom to upload custom models/nodes
  • 50+ ready-to-run workflows
  • 100% private workspace with up to 200GB storage
  • Dedicated Support

Run ComfyUI Online

PuLID_ComfyUI Description

PuLID_ComfyUI is a native implementation of the PuLID framework within ComfyUI, designed to enhance user interface functionality and integration. It streamlines UI processes, offering improved performance and user experience.

PuLID_ComfyUI Introduction

PuLID_ComfyUI is a native implementation of the PuLID (Pure and Lightning ID Customization via Contrastive Alignment) model within the ComfyUI framework. This extension is designed to help AI artists generate high-quality, customized images by leveraging advanced techniques in image generation and contrastive alignment. PuLID_ComfyUI allows for fine-tuning and precise control over the generated images, making it an invaluable tool for artists looking to create unique and detailed visuals.

How PuLID_ComfyUI Works

PuLID_ComfyUI operates by applying weights to different aspects of the image generation process, allowing for customization in terms of fidelity to the reference image and stylistic freedom. The extension uses a 4-step lighting UNet model to ensure high-quality outputs. By adjusting parameters such as fidelity and projection, users can control how closely the generated image resembles the reference image or how much creative freedom is allowed.

Basic Principles:

  • Fidelity vs. Style: The method parameter lets you choose between higher fidelity to the reference image or more stylistic freedom. Higher fidelity means the generated image will closely match the reference, while more stylistic freedom allows for creative variations.
  • Advanced Node: This feature provides additional control with a fidelity slider and projection options, enabling more nuanced adjustments.

PuLID_ComfyUI Features

Attention Masking

This feature allows for fine-tuning specific areas of the image, ensuring that important details are preserved or enhanced. Attention masking can be particularly useful when working with complex images that require detailed customization.

Advanced Node

The Advanced node offers more granular control over the image generation process. It includes:

  • Fidelity Slider: Adjusts the level of detail and resemblance to the reference image.
  • Projection Options: Different projection methods (ortho_v2, ortho) that affect the stylistic outcome of the image.

Method Parameter

The method parameter provides three options:

  • Fidelity: Prioritizes resemblance to the reference image.
  • Style: Allows for more creative freedom and stylistic variations.
  • Neutral: No normalization, offering a balanced approach.

PuLID_ComfyUI Models

PuLID_ComfyUI primarily uses the 4-step lighting UNet model, which is known for its high-quality outputs. The model ensures that the generated images are both detailed and visually appealing. Users can experiment with different models and settings to achieve the desired results.

Model Variations:

  • Standard Model: Best for general use, providing a good balance between fidelity and style.
  • Lightning Lora: May not perform as well in some cases, but can be useful for specific applications.

What's New with PuLID_ComfyUI


  • 2024.05.12: Added attention masking and the Advanced node, allowing for more precise control over the image generation process. These updates enhance the flexibility and functionality of PuLID_ComfyUI, making it easier for AI artists to achieve their creative goals.

Troubleshooting PuLID_ComfyUI

Common Issues and Solutions:

  1. Quality Degradation with Other Models:
  • Solution: Experiment with different CFG (Classifier-Free Guidance) settings and samplers/schedulers like sgm_uniform.
  1. Reference Image Quality:
  • Solution: Ensure that the reference image is clean and sharp, as the quality of the reference image significantly impacts the final output.
  1. IPAdapter Compatibility:
  • Solution: Update the IPAdapter extension to ensure compatibility with PuLID_ComfyUI.

Frequently Asked Questions:

  • Q: Why is my generated image not matching the reference image closely?
  • A: Try increasing the fidelity value or using the fidelity method in the Advanced node.
  • Q: How can I achieve more stylistic freedom in my images?
  • A: Use the style method or lower the fidelity value in the Advanced node.

Learn More about PuLID_ComfyUI

For additional resources, tutorials, and community support, you can explore the following links:

  • These resources provide comprehensive information and support to help you make the most of PuLID_ComfyUI.

PuLID_ComfyUI Related Nodes


© Copyright 2024 RunComfy. All Rights Reserved.

RunComfy is the premier ComfyUI platform, offering ComfyUI online environment and services, along with ComfyUI workflows featuring stunning visuals.