ComfyUI  >  Nodes  >  ComfyUI-TCD

ComfyUI Extension: ComfyUI-TCD

Repo Name

ComfyUI-TCD

Author
JettHu (Account age: 2355 days)
Nodes
View all nodes (1)
Latest Updated
6/3/2024
Github Stars
0.1K

How to Install ComfyUI-TCD

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

ComfyUI-TCD Description

ComfyUI-TCD integrates ComfyUI with the TCD framework, enhancing user interface capabilities by leveraging TCD's features. This extension streamlines UI development, offering a seamless experience for developers.

ComfyUI-TCD Introduction

ComfyUI-TCD is an extension for the ComfyUI platform that implements the TCD (Trajectory Consistency Distillation) sampler. This extension is inspired by Consistency Models and Latent Consistency Models (LCM) and offers a novel approach to distilling pre-trained diffusion models. The primary advantage of using TCD is its ability to generate high-quality, detailed images with fewer denoising steps compared to traditional methods like LCM. This makes it particularly useful for AI artists looking to create detailed and high-quality images efficiently.

How ComfyUI-TCD Works

ComfyUI-TCD works by distilling a pre-trained diffusion model into a more efficient model that can produce high-quality images in fewer steps. The process involves reducing the number of denoising steps required to generate an image, which speeds up the image creation process without sacrificing quality. TCD achieves this by controlling the strength of random noise during the denoising process, allowing for fine-tuning of image details.

Basic Principles

  1. Distillation: TCD distills a pre-trained diffusion model into a more efficient form.
  2. Denoising Steps: The number of steps required to denoise an image is reduced, making the process faster.
  3. Noise Control: Parameters can be adjusted to control the amount of random noise, affecting the level of detail in the final image.

ComfyUI-TCD Features

Detailed Image Generation

  • High-Quality Output: TCD generates images with more details compared to LCM, even with fewer denoising steps.
  • Adjustable Parameters: Users can control the level of detail by adjusting parameters like eta, which controls the strength of random noise.

Customizable Settings

  • Steps: Number of denoising steps, similar to the steps parameter in the KSampler node.
  • Scheduler: Sampling scheduler options include simple and sgm_uniform. The simple scheduler generates results similar to diffusers, while sgm_uniform is recommended by the ComfyUI author.
  • Denoise: Controls the denoising strength, similar to the denoise parameter in the KSampler node.
  • Eta: Controls the strength of random noise during denoising. A higher eta value results in more randomness and potentially more detailed images.

ComfyUI-TCD Models

ComfyUI-TCD supports various models, including:

  • TCD-SD15-LoRA: Suitable for SDv1.5 models.
  • TCD-SDXL-LoRA: Suitable for SDXL models.
  • Hyper-SD15-1step-lora: Optimized for 1-step generation with SDv1.5. - Hyper-SDXL-1step-lora: Optimized for 1-step generation with SDXL. These models can be used to achieve different levels of detail and performance, depending on the specific requirements of your project.

What's New with ComfyUI-TCD

Updates

  • 2024.4.28: Official PR submitted to ComfyUI.
  • 2024.4.28: Initial release of the repository. These updates bring improved performance and new features, making it easier for AI artists to create high-quality images efficiently.

Troubleshooting ComfyUI-TCD

Common Issues and Solutions

  1. Image Quality Issues: If the generated images lack detail, try adjusting the eta parameter to control the level of random noise.
  2. Performance Problems: Ensure that your system meets the minimum requirements for running ComfyUI and the TCD extension.
  3. Scheduler Errors: If you encounter issues with the scheduler, try switching between simple and sgm_uniform to see which works best for your setup.

Frequently Asked Questions

  • Q: How do I adjust the level of detail in my images?
  • A: Use the eta parameter to control the strength of random noise during denoising. Higher values result in more detailed images.
  • Q: What are the recommended settings for high-quality images?
  • A: Start with an eta value of 0.3 and adjust based on your specific needs. Experiment with different steps and scheduler settings to find the best combination.

Learn More about ComfyUI-TCD

For additional resources, tutorials, and community support, consider exploring the following:

  • : The official repository for the ComfyUI-TCD extension.
  • : Learn more about the theoretical background and development of TCD.
  • : Join the community to ask questions, share your work, and get support from other AI artists and developers. By leveraging these resources, you can enhance your understanding of ComfyUI-TCD and make the most of its powerful features for your AI art projects.

ComfyUI-TCD Related Nodes

RunComfy

© 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.