ComfyUI - Bắt đầu : Tập 1 - Tốt hơn AUTO1111 dành cho nghệ thuật tạo ảnh Stable Diffusion

Trung Caha
6 Aug 202324:15

TLDRThe video script introduces viewers to the latest stable defusion (sdxl) version in a comprehensive tutorial. The host, Trung, guides users through the interface, explaining how to create a workfollow chain to generate images. The tutorial covers model selection, command scripting, image sizing, and the use of various features like the 'clip take and colder' command. It also details the process of refining images for better quality using advanced settings and concludes with sharing options for the created workflow. The video aims to demystify the platform and empower users to create high-quality images with ease.


  • 📝 The video is a tutorial on using the latest stable defusion version (sdxl) with a new interface called 'compai'.
  • 🚀 The tutorial guides users through creating a workflow to generate images, making it accessible for beginners and avoiding overwhelming instructions.
  • 🔍 The presenter provides an overview of the interface and its features, including model versions, command inputs, and image sizing.
  • 🎨 Users can customize the appearance of the interface, such as changing colors and titles for better visibility and organization.
  • 🖌️ The process involves writing commands in designated areas and connecting them to build the image generation sequence.
  • 🌐 The tutorial emphasizes the ease of dragging and dropping elements to create connections between different parts of the workflow.
  • 📸 The 'preview' feature allows users to review and save the generated images, with options to adjust the image quality and settings.
  • 🔄 The presenter demonstrates how to create a second, improved image by adjusting parameters and using different image generation steps.
  • ⚙️ The importance of connecting all elements correctly to avoid errors, such as the red error sign indicating a missing connection, is highlighted.
  • 📈 The video showcases the potential for image quality improvement by comparing the original and enhanced images.
  • 💡 The presenter encourages viewers to share their workflows and provides instructions on how to save and share the 'compai' workflow with others.

Q & A

  • What is the main topic of the video?

    -The main topic of the video is a tutorial on how to use the stable defusen version of the software with a new interface called compai.

  • What is the first step in accessing the stable defusen version of the software?

    -The first step is to log in to the stable defusen interface, which is demonstrated in a previous video mentioned by the speaker.

  • What does the 'clear' button in the interface do?

    -The 'clear' button resets the interface, allowing users to start fresh and build their workstreams from scratch.

  • What is the purpose of the first box in the workstream?

    -The first box is where users will select the model versions they need to create their images.

  • How can users adjust the size of the images they create?

    -Users can adjust the image size by selecting the 'resolution' option and inputting their desired dimensions, such as 512x512 or 1024x1024.

  • What is the role of the 'QR Prime' section in the workflow?

    -The 'QR Prime' section is where users will write their commands to create the images. It is crucial for specifying the details of the image generation process.

  • What are the two main commands needed in the 'Steven' part of the workflow?

    -The two main commands needed are 'passive Prime' for the areas of the image the user wants to create, and 'chip' for the elements that the user does not want to appear in the image.

  • How can users change the color of the boxes in the interface?

    -Users can change the color of the boxes by right-clicking, selecting 'color', and then choosing a new color from the available options.

  • What is the purpose of the 'preview' button in the final part of the workflow?

    -The 'preview' button allows users to view their image before it is fully processed and saved, giving them a chance to review and make adjustments if needed.

  • How can users save their workflow for future use or share it with others?

    -Users can save their workflow by clicking 'save', which will store it as a JSON file. This file can then be shared with others or used again later by the user.

  • What is the significance of the 'upscale' process demonstrated in the second part of the video?

    -The 'upscale' process is used to improve the quality of the generated image by increasing its resolution and detail. This is done by using advanced settings and additional steps in the workflow.



🚀 Introduction to Stable Diffusion SDL Version

The video begins with an introduction to the Stable Diffusion SDL version by the host, Trung. He explains that he will guide the viewers through a new series focused on utilizing this version with a user-friendly interface. The primary goal is to help the audience understand how to create workstreams effectively without being overwhelmed by complex instructions. Trung provides a walkthrough on accessing and navigating the Stable Diffusion SDL interface, emphasizing its ease of use and intuitive design.


📝 Customizing the Workflow and Commands

In this segment, Trung delves into the customization of the workflow and commands within the Stable Diffusion SDL interface. He demonstrates how to modify the title and various settings to suit the user's preferences. The host also explains the process of connecting and using different commands, such as 'clip take and colder,' to refine the image generation. The focus is on making the interface more recognizable and visually appealing by changing colors and organizing the workflow efficiently.


🖼️ Image Generation and Parameters

This paragraph covers the crucial steps in image generation using Stable Diffusion SDL. Trung introduces the concept of 'lation emich' and explains its role in setting the image size according to the user's requirements. He details the process of selecting the number of images to generate and the parameters that control the image quality. The host also discusses the importance of connecting the right commands and parameters to achieve the desired outcome, emphasizing the simplicity and efficiency of the Stable Diffusion SDL interface.


🔍 Reviewing and Saving the Image

The fourth paragraph focuses on the final stages of image generation and review. Trung explains how to preview and save the generated images, highlighting the ease of transitioning from the command code to the final visual product. He also discusses the importance of the 'Email' section, where users can review their images or save them directly. The host emphasizes the user-friendly nature of the Stable Diffusion SDL interface, making it accessible for users to manage their image generation projects effectively.


🌟 Enhancing Image Quality and Sharing Workflows

In the concluding paragraph, Trung demonstrates how to enhance the quality of the generated images by using advanced features and settings within the Stable Diffusion SDL interface. He shows the process of creating a second, improved workflow to double the image quality and explains the importance of connecting the right models and parameters for this enhancement. The host also talks about sharing workflows with others, either by saving the file or opening the image in a new tab. The video ends with a call to action for viewers to share their feedback and experiences with the new Stable Diffusion SDL interface.



💡stable defusen

The term 'stable defusen' refers to a specific version of a software or tool used for image generation and manipulation. In the context of the video, it is the primary tool being discussed and demonstrated, with the presenter guiding viewers on how to use it effectively. The script mentions selecting different versions of 'stable defusen' and using it to create images, indicating its importance in the video's theme.


The 'interface' in this context refers to the visual and interactive elements of the software 'stable defusen' that users interact with to perform tasks. The video emphasizes the importance of understanding and navigating the interface to create images effectively. It is a key concept as it is the medium through which users engage with the software.


In the context of the video, 'workfollow' seems to refer to the sequence of steps or the workflow that users need to follow to create images using 'stable defusen'. It is a critical concept as it outlines the process from start to finish, guiding users on how to use the software effectively.


Commands are specific instructions or inputs that the user writes to control the 'stable defusen' software and direct the image creation process. Understanding and using the correct commands is essential for achieving the desired output in the software.

💡image generation

Image generation is the process of creating new images using the 'stable defusen' software. It involves a series of steps, including selecting models, writing commands, and setting parameters to produce the final visual output. This is the main goal and outcome of following the 'workfollow' process.


Parameters are the specific settings or values that the user can adjust within the 'stable defusen' software to influence the image generation process. These can include the size of the image, the number of images to generate, and other qualities that determine the final product.


Previewing in the context of the video refers to the ability to view the image before it is fully rendered or saved. This feature allows users to assess the quality and accuracy of the image generation process and make adjustments if necessary.

💡save image

Saving an image is the action of preserving the generated image to a user's device or a specified location. This is the final step in the image generation process, ensuring that the user can retain and use the created image for their purposes.


A workflow in this context refers to the sequence of operations or steps that the user follows to accomplish a task within the 'stable defusen' software. It encompasses the entire process from starting the image generation to saving the final product.

💡quality improvement

Quality improvement refers to the enhancement of the image's visual attributes, such as resolution, detail, and overall appearance. In the video, the user seeks to improve the quality of the generated images by adjusting settings and using different methods within the 'stable defusen' software.


Introduction to the latest series on using stable defultion version sdxl with a new interface called compai.

Explanation of how to access and log in to the stable defusion version sdxl.

Overview of the new interface and its basic functions.

Step-by-step guide on constructing a workfollow chain to create images using the new interface.

Details on selecting and using different model versions for image creation.

Instructions on writing commands and using connected fields for image generation.

Demonstration of resizing and zooming in on images for better viewing.

Selection of desired model versions and customization of interface colors for better visibility.

Process of creating a new chain for image generation with specific commands.

Explanation of how to use the 'clip take and colder' command for image creation.

Guidance on changing the title and color of different sections for better organization.

Description of copying and pasting techniques within the compai interface.

Discussion on setting up the image size and generation parameters.

Explanation of the 'negate' function and its role in image creation.

Process of creating a new, improved version of an image by adjusting parameters and using advanced features.

Demonstration of the image generation process and its dependence on computer configuration and image size.

Review of the entire workflow and its simplicity in creating images on the compai interface.

Comparison of the quality between a basic image and an upscaled image to showcase the effectiveness of the compai interface.

Instructions on how to save and share the created workflow with others.