This One Simple Plugin Adds Realtime AI Assistance to Krita

Nerdy Rodent
24 Nov 202314:31

TLDRThe video script introduces viewers to the world of live stable diffusion, a technique that allows for the creation of masterpieces in almost real time. It guides users through the installation and setup process of necessary software and extensions, emphasizing the ease of use for both beginners and experienced users. The script highlights the capabilities of the system, such as real-time image generation and manipulation, and the integration of control nets for dynamic posing and character adjustments. The tutorial is engaging, informative, and encourages users to explore the creative potential of stable diffusion technology.


  • 🎨 The script introduces a method to draw an owl using simple shapes, highlighting the creative process.
  • πŸš€ The video focuses on using live stable diffusion to create masterpieces in real time with the aid of a powerful UI.
  • 🌐 LCM is used in the process, and having a comfy UI is beneficial as it simplifies the setup.
  • πŸ’» System requirements for stable diffusion include at least 6GB of VRAM and compatibility with Linux, Windows, or experimental Mac support.
  • πŸ”§ Installation of Creer is necessary, with instructions provided for both Linux and Windows users.
  • πŸ“‚ The script explains the process of downloading and installing the AI image diffusion plugin into the resources folder.
  • πŸ”„ Restarts of the Creer application are required to enable the new plugin and configure settings.
  • πŸ–ŒοΈ The user can generate images in real time, with options to adjust brush size, noise strength, and seed for randomness.
  • πŸ“Έ The video demonstrates how to use control nets and pose adjustments to refine the generated images.
  • 🎭 The script emphasizes the versatility of the tool, allowing for both freeform scribbling and detailed, prompt-based image generation.
  • πŸ› οΈ Troubleshooting tips are provided for issues such as connection errors, guiding users to check log files and ensure proper installation of nodes and models.

Q & A

  • What is the main topic of the video?

    -The main topic of the video is about using live stable diffusion in conjunction with Creer and Comfy UI to create images in real time.

  • What are the system requirements to run stable diffusion?

    -The system requirements to run stable diffusion include a computer with at least 6 gigabytes of VRAM and an operating system such as Linux or Microsoft Windows. Experimental support for Mac OS is also available.

  • What is Creer and why is it needed for this process?

    -Creer is a painting application that provides various brushes and tools, similar to other painting packages. It is needed because the stable diffusion extension will be installed and run within the Creer environment.

  • How can one install Creer?

    -Creer can be installed via a single click in the free software store on Linux or through the Creer website.

  • What version of Creer is recommended for this process?

    -The recommended version of Creer for this process is 5.2.1, as mentioned in the script.

  • What is the role of the resources folder in the process?

    -The resources folder is where the plugin for stable diffusion is unzipped and installed. It is found within the Creer directory.

  • How does one enable the AI image diffusion plugin?

    -To enable the AI image diffusion plugin, one needs to go to the settings of Creer, find the 'Python plugin manager' section, tick the box for 'AI image diffusion', and then restart Creer.

  • What are the two options for managing the local server?

    -The two options for managing the local server are 'Local server managed by the Creer plugin' or 'Connect to external server (local or remote)'.

  • What are the required custom nodes for the stable diffusion plugin?

    -The required custom nodes for the stable diffusion plugin include Control Net, Pre-processors, IP adapter, Ultimate SD upscale, and External tooling nodes.

  • How can one troubleshoot issues with the model not being found?

    -If the model is not being found, one can refer to the troubleshooting section in the GitHub page or open the client.log and server.log files to get more information about what is happening.

  • What can one do with the live mode in the stable diffusion plugin?

    -In the live mode, one can draw and the stable diffusion will interpret the drawing in real time. The user can adjust the strength of the noise, add control layers, and manipulate the generated image with various settings.



🎨 Introduction to Live Stable Diffusion and Creer Setup

This paragraph introduces the audience to the process of setting up live stable diffusion using Creer, a powerful tool that allows users to create masterpieces in real time. It emphasizes the simplicity of the setup process, requiring minimal resources and compatibility with various operating systems. The speaker also highlights the benefits of Creer, such as the availability of different brushes and tools, and the ability to generate images with minimal effort. The paragraph outlines the initial steps for installation and configuration, including the need for a computer with sufficient VRAM, the installation of Creer, and the extraction of the plugin into the resources folder. It also touches on the importance of ensuring that the computer meets the requirements to run stable diffusion and provides guidance on how to proceed with the setup.


πŸ”§ Advanced Setup with Optional Custom Comfy UI Server

This paragraph delves into the optional setup of a custom Comfy UI server for users who are already familiar with the process and have the necessary extensions and models. It addresses the issue of connection failure due to the lack of Comfy UI setup and provides a link to the GitHub page for required extensions and models. The speaker explains the need for specific custom nodes and models, and suggests using the Comfy UI manager for easy installation. The paragraph also discusses the handling of model names and directories, and offers troubleshooting tips for issues with model detection. It then explains how to establish a connection with the server once Comfy UI is up and running, and briefly mentions additional options available in the menu for changing models and adjusting settings.


πŸ–ŒοΈ Real-Time Drawing and Image Generation

This paragraph demonstrates the real-time drawing capabilities of the stable diffusion plugin, showcasing how users can create and manipulate images live. The speaker guides the audience through resizing the brush, drawing a character, and adjusting the 'strength' to achieve a desired artistic effect. It also highlights the ability to add and manipulate vector layers and control character poses, as well as the option to freely scribble without prompts. The paragraph emphasizes the fun and creative potential of the tool, and provides a practical example of how to use it to generate and modify an image. It concludes with a mention of future videos that will cover additional features and integration of existing images into the stable diffusion art.



πŸ’‘Stable Diffusion

Stable Diffusion is an advanced AI model that generates images from textual descriptions. It is a form of deep learning that uses a process called diffusion to create high-quality, realistic images. In the video, the presenter is discussing how to utilize Stable Diffusion within the Creer application to generate images in real-time, showcasing its capabilities in transforming simple sketches into detailed illustrations.

πŸ’‘Comfy UI

Comfy UI refers to a user-friendly graphical user interface that simplifies the process of interacting with complex software, in this case, for AI image generation. The script emphasizes the importance of having Comfy UI installed to fully benefit from the features and capabilities of the AI image diffusion plugin in Creer. It is a key component for managing models and custom nodes for the AI image generation process.

πŸ’‘Live Mode

Live Mode is a feature within the Creer application that allows users to draw and have the AI model, Stable Diffusion, generate images in real-time based on their sketches. This interactive mode provides instant feedback and allows for dynamic adjustments, such as changing the 'strength' of the AI's influence or adding elements like control nets for posing characters.

πŸ’‘Control Nets

Control Nets are a set of predefined vectors or guidelines within the Creer application that help in manipulating and posing characters in a specific way. They are a part of the Comfy UI system and are used in conjunction with the AI image generation to add dynamic elements, such as character poses, to the generated images.

πŸ’‘AI Image Diffusion Plugin

The AI Image Diffusion Plugin is an extension for the Creer application that integrates the Stable Diffusion AI model, enabling users to generate images directly within the Creer environment. This plugin allows for a seamless workflow between drawing and AI-assisted image creation, offering features like real-time drawing interpretation and image generation from textual prompts.


Docker is a platform that enables developers to deploy applications by using containers. In the context of the video, Docker is used to manage the local server that runs the Stable Diffusion model. The script explains how to configure Docker within the Creer application to either manage it through the plugin or connect to an external server.

πŸ’‘Vector Layer

A Vector Layer in the context of the Creer application is a layer that contains vector graphics, which are scalable and resolution-independent images. Vector layers are used to add control elements, such as character poses, to the artwork. The script describes using a vector layer to add a control layer for character poses, which can then be manipulated in real-time.

πŸ’‘Custom Nodes

Custom Nodes in the context of the video refer to specific extensions or modules that are required for the Comfy UI to function properly with the AI image diffusion plugin. These nodes enhance the capabilities of the Creer application by providing additional features and functionalities needed for the AI image generation process.

πŸ’‘Model Upscale

Model Upscale refers to the process of increasing the resolution or quality of a generated AI model without losing detail or clarity. In the context of the video, it is one of the required models for the AI image diffusion plugin. The script mentions an upscaler that is used to enhance the quality of the generated images.


In the context of AI image generation, a Prompt is a textual description or input that guides the AI model in creating an image. The quality and specificity of the prompt can significantly influence the output of the generated image. The video script describes how users can enter prompts in Live Mode to direct the AI in interpreting their sketches.

πŸ’‘Real-time Drawing

Real-time Drawing refers to the ability of the Creer application, with the help of the AI image diffusion plugin, to instantly interpret and visualize a user's sketch as they draw. This feature allows for immediate feedback and adjustments, making the creative process more interactive and dynamic.


The ability to draw an owl in two simple steps: starting with circles and then adding the rest of the features.

Introduction to live stable diffusion, a technique for creating masterpieces in almost real time.

The necessity of a computer with at least 6 gig for VRAM to run stable diffusion smoothly.

Support for various operating systems including Linux, Microsoft Windows, and experimental support for Mac.

The requirement of having Creer installed, which can be done through the free software store on Linux or via the Creer website.

The current recommended version for Creer is 5.2.1.

Instructions on how to configure Creer settings and where to unzip the plugin.

Enabling the plugin through the Python plugin manager in Creer's settings.

The option to either have a local server managed by the plugin or to manage your own local server.

The need to install additional extensions and models if not already done, with a guide on how to do so using the Creer UI manager.

Troubleshooting tips such as checking the client and server log files for errors.

Exploration of the St menu and its various options for changing models, prompts, and other settings.

The live drawing feature, which allows users to draw and have the image generated in real time.

The ability to adjust the strength of the noise and the seed for randomization in the live drawing mode.

The integration of control nets and the use of vector layers for dynamic image manipulation.

The interpretation of free-form scribbling without prompts, showcasing the AI's capability to generate images from abstract input.

The potential for users to copy and paste generated images for further editing and customization.

The mention of a subsequent video that will demonstrate how to integrate existing images into stable diffusion art.