How to install and use ComfyUI - Stable Diffusion.

Sebastian Kamph
14 Jul 202312:44

TLDRIn this informative video, the creator guides viewers through the installation and basic usage of Comfy UI, a powerful and flexible user interface that allows for the creation of custom workflows. The tutorial covers downloading the software from GitHub, extracting the files, and setting up the interface with various models and nodes. It also discusses troubleshooting tips, updating the software, and utilizing the Comfy UI manager for installing custom nodes. The video showcases the potential of node-based systems for image generation and provides examples of different workflows, encouraging users to explore and learn from others' creations.


  • πŸ˜€ Comfy UI offers total freedom and control for creating custom workflows, enhancing user experience in developing user interfaces.
  • πŸš€ The installation process for Comfy UI is straightforward, involving downloading from GitHub, extracting files with 7-Zip or WinRAR, and following setup instructions for GPU or CPU use.
  • πŸ”§ Advanced configuration options allow users to integrate Comfy UI with existing models from Automatic 1111 or Stable Fusion, streamlining the workflow setup.
  • 🌐 For users without pre-existing models, downloading a model from Civitai is recommended to start utilizing Comfy UI effectively.
  • 🎨 The interface of Comfy UI is node-based, allowing for the flexible connection of features and creation of complex workflows through drag-and-drop functionality.
  • πŸ”„ Users can easily update Comfy UI and add advanced extensions for enhanced capabilities, following guidance in the README documentation.
  • πŸ’‘ Negative and positive prompts can be managed and adjusted within the UI, illustrating the model's versatility in image generation and customization.
  • πŸ“ Troubleshooting tips and model checkpoint updates are readily available, ensuring users can maintain and enhance their Comfy UI setup.
  • πŸ” A unique feature of Comfy UI is the ability to import workflows from other users, enabling learning and experimentation with different settings and nodes.
  • 🎞️ Comfy UI supports a wide array of nodes and customization options, from basic setup to advanced configurations for experienced users.

Q & A

  • What is Comfy UI?

    -Comfy UI is a powerful user interface that provides total freedom and control for users to create their own workflows.

  • How can you install Comfy UI?

    -To install Comfy UI, visit GitHub, find the Comfy UI repository, and download the .zip file for your operating system. Extract the file to your desired location and follow the instructions in the README.

  • What is the significance of the Nvidia GPU in Comfy UI?

    -The Nvidia GPU is recommended for running Comfy UI as it provides better performance compared to running on CPU. The README file provides specific instructions for users with Nvidia GPUs.

  • What is the purpose of the model checkpoint in Comfy UI?

    -The model checkpoint is required for the proper functioning of Comfy UI. It ensures that the system can access the necessary models for generating images.

  • How can you update Comfy UI if you already have models installed?

    -If you have models installed from another version of Comfy UI or Stable Fusion, you can update by accessing the 'update' option in the config UI directory.

  • What is the role of nodes in Comfy UI?

    -Nodes in Comfy UI are used to connect different features and create custom workflows. They allow users to control how features interact with each other and the overall image generation process.

  • How do you select a model in Comfy UI?

    -In Comfy UI, you select a model by using the 'Load Checkpoint' or 'Load Models' node and choosing the desired model from the available options.

  • What is the purpose of the positive and negative prompts in Comfy UI?

    -Positive and negative prompts in Comfy UI are text inputs that guide the AI in generating images. The positive prompt describes the desired features, while the negative prompt specifies what should not be included.

  • How can you customize the image generation process in Comfy UI?

    -You can customize the image generation process in Comfy UI by adjusting settings such as the seed, which determines the base noise for image generation, and the number of steps the image will go through while rendering.

  • What is the Comfy UI Manager and how does it help with custom nodes?

    -The Comfy UI Manager is a tool that helps users install and manage custom nodes. It allows users to add new nodes to their Comfy UI setup, enhancing the capabilities and flexibility of their workflows.

  • How can you learn more about advanced features and workflows in Comfy UI?

    -For more advanced features and workflows, users can refer to the examples and tutorials provided in the Comfy UI's GitHub repository, which includes a variety of node setups and use cases.



πŸš€ Introduction to Comfy UI Installation and Basics

This paragraph introduces the Comfy UI, a powerful user interface that offers complete freedom and control in creating workflows. It outlines the process of installing Comfy UI, including navigating to GitHub, downloading the necessary files, and extracting them. The speaker also mentions an advanced extension and provides troubleshooting tips for common issues like a red error in the UI. The importance of having an Nvidia GPU for smoother operation is highlighted, and instructions are given for users with or without an Nvidia GPU. The paragraph concludes with a brief mention of the readme file's significance and the next steps for users with existing installations of other models.


πŸ“Œ Advanced Workflows and Customization

The second paragraph delves into the advanced features of Comfy UI, focusing on the node-based system that allows users to connect features and create custom workflows. It explains the process of loading models, setting up prompts, and using the sampler to render images. The paragraph also touches on the configuration settings for image generation, such as width, height, batch size, and seed. The concept of the K sampler and the CFG scale is introduced, along with recommendations for settings based on the stable Fusion 1.5 and 2.0 models. The paragraph concludes with a discussion on adding more nodes and the availability of various nodes for further customization.


🌐 Importing Workflows and Using the Comfy UI Manager

This paragraph discusses the ability to import workflows from other users' images and the power of learning from existing workflows. It introduces the Comfy UI manager, which is essential for installing custom nodes that might be missing. The process of downloading the manager, extracting its contents, and restarting Comfy UI to enable the installation of custom nodes is detailed. The paragraph also showcases how to use the manager to fix errors by installing missing custom nodes. It concludes with an example of how the system can replicate a user's workflow by loading an image and emphasizes the flexibility and control offered by the node-based system of Comfy UI.



πŸ’‘Comfy UI

Comfy UI is a powerful user interface that provides users with the freedom and control to create their own workflows. It is a node-based system that allows for the connection and customization of various features, enabling users to build complex image generation processes. In the context of the video, Comfy UI is the main focus, and the tutorial walks users through its installation and basic usage, highlighting its flexibility and potential for advanced image rendering.


Workflows refer to the series of steps or processes that are followed in Comfy UI to achieve a specific outcome, such as generating an image. These are created by connecting various nodes in the interface, each representing a different function or operation. The video provides advanced workflows as examples and also explains how users can create their own by connecting different nodes according to their preferences or requirements.


Nodes in the context of Comfy UI are the building blocks of a workflow. They represent individual components or functions that can be connected to create a complete process. Nodes can handle tasks such as loading models, processing text prompts, and rendering images. The power of nodes lies in their ability to be interconnected in various ways, allowing for a high degree of customization and control over the image generation process.


GitHub is a web-based platform that provides version control and collaboration features for software development. In the video, GitHub is mentioned as the source for downloading the Comfy UI software. It is a central hub where developers can share code, track changes, and collaborate on projects, making it an essential tool for open-source development and community-driven projects.

πŸ’‘Nvidia GPU

Nvidia GPU refers to the Graphics Processing Unit (GPU) manufactured by Nvidia, a company known for its high-performance graphics and computing solutions. GPUs are used in Comfy UI to accelerate the rendering process of images, providing faster and more efficient computation. The video script mentions that users with an Nvidia GPU should follow a specific path file for optimal performance.


7z is a file format and the name of the free and open-source file archiver software that uses this format. It is known for its high compression ratio, making it suitable for reducing the size of files for storage or transfer. In the context of the video, 7z is used to extract the downloaded Comfy UI file, which is a large archive that needs to be decompressed before use.

πŸ’‘Model Checkpoint

A model checkpoint refers to a point in the training process of a machine learning model where the model's state is saved. This saved state can be used to resume training or for inference at a later time. In the context of Comfy UI, a model checkpoint is a file that contains the trained parameters of a neural network, which is used to generate images based on input prompts.

πŸ’‘Config UI

Config UI refers to the user interface within Comfy UI that allows users to configure settings and preferences for their workflows. This includes setting the base path for models, renaming example folders, and adjusting other parameters that affect how the UI operates and interacts with different models.

πŸ’‘Custom Nodes

Custom nodes are additional components that can be installed in Comfy UI to expand its functionality and provide more options for users in creating workflows. These nodes are not part of the standard installation and must be downloaded and installed separately, often to support specific features or models.

πŸ’‘Comfy UI Manager

The Comfy UI Manager is a tool designed to simplify the installation and management of custom nodes in Comfy UI. It provides users with an easy way to install missing custom nodes or update existing ones, ensuring that the UI has all the necessary components to run complex workflows and render images as intended.


In-painting is a technique used in image editing where missing or unwanted parts of an image are filled in or 'painted' with new content that matches the surrounding area. In the context of Comfy UI, an in-painting node is used to generate content for specific areas of an image, such as adding details to an existing picture or covering up certain elements.


Comfy UI is a powerful user interface that offers total freedom and control to create custom workflows.

The installation process for Comfy UI involves downloading from GitHub and extracting the .7z file using a program like WinRAR or 7-Zip.

Users with Nvidia GPUs should follow specific instructions outlined in the README file for optimal performance.

Comfy UI offers an advanced extension that can be utilized for more sophisticated workflows.

The UI features a node-based system where users can connect various features and create personalized workflows.

Users can download models from Civitai to enhance their Comfy UI experience if they do not have any models installed.

Comfy UI allows users to import workflows from other users' images, providing a valuable learning resource and creative inspiration.

The Comfy UI Manager can be used to install custom nodes, ensuring that users have access to all necessary components for their desired workflows.

Comfy UI provides a default interface with nodes for loading models, setting prompts, and rendering images.

The node-based system in Comfy UI enables users to switch around elements and create unique combinations for image generation.

Comfy UI includes a variety of nodes, allowing users to add more functionality and customization to their workflows.

The UI provides examples and tutorials on GitHub for users to learn from, including setups for tasks like in-painting and advanced sampling techniques.

Comfy UI's node system allows for the adjustment of parameters such as the CFG scale, which influences the AI's responsiveness to prompts.

Users can rename and reorganize nodes in Comfy UI to better understand and control their workflows.

Comfy UI supports the use of custom nodes, which can be installed through the Comfy UI Manager, expanding the capabilities of the UI.

The tutorial provides a step-by-step guide on setting up and using Comfy UI, including troubleshooting tips for common issues.

Comfy UI's interface includes options for setting the image resolution, batch size, and seed for image generation.