How To Install Invoke AI on Windows in 2023

Bill Meeks
19 Nov 202313:32

TLDRThis tutorial offers a step-by-step guide on installing Invoke AI on Windows for professional artists and designers. It covers the technical requirements, including an Nvidia graphics card with at least 4 GB of VRAM, 12 GB of RAM, and 18 GB of free hard drive space. The process involves downloading the installer from Invoke AI's GitHub page, extracting it, and running the install.bat file. The video also explains how to configure the installation manually, including setting up the GPU type, downloading models from Huggingface, and selecting the desired models for installation. After installation, users can add existing stable diffusion models to Invoke AI and explore its features. The tutorial concludes with a demonstration of generating images using Invoke AI, showcasing its capabilities for creative professionals.

Takeaways

  • ๐ŸŽจ **Invoke AI Introduction**: Invoke AI is a generative AI tool designed for professional artists and designers, combining stable diffusion with professional workflows and a clean, intuitive UI.
  • ๐Ÿ’ก **Technical Requirements**: To run Invoke AI on Windows, you need an Nvidia graphics card with at least 4 GB of VRAM (8 GB for the latest stable diffusion version), 12 GB of RAM, and at least 18 GB of free hard drive space.
  • ๐Ÿ“ฅ **Software Dependencies**: Ensure you have the correct version of Python (3.10 or 3.11) and the Microsoft Visual C++ redistributable from 2015 to 2022 installed.
  • ๐Ÿ”— **Installation Process**: The installation involves downloading the Invoke AI installer from GitHub, extracting it, and running the install.bat file.
  • ๐Ÿ“ **Folder Setup**: You can choose to install Invoke AI in a custom folder to keep it off your main drive.
  • ๐Ÿš€ **GPU Selection**: The installer will prompt you to select the type of GPU you have, with Nvidia GPUs being the only option for Windows users.
  • ๐Ÿ”„ **Configuration Options**: You can choose between an automatic configuration that gets base models or a manual configuration for more control over the models and settings.
  • ๐Ÿ“š **Model Management**: After installation, you can add existing stable diffusion models to Invoke AI through the model manager.
  • ๐Ÿ–ผ๏ธ **Image Output**: Invoke AI allows you to save images to a separate folder, which can be linked to a database for cataloging your creations.
  • ๐Ÿ” **Model Importing**: You may encounter issues when importing some models, but the process should allow for troubleshooting and resolution.
  • ๐ŸŒ **Web Interface**: Invoke AI can be launched through a browser-based interface by running the invoke dobat file and following the provided link.
  • โœ… **Testing the Setup**: To confirm everything is working, you can run test prompts through Invoke AI and observe the generation process in the command window.

Q & A

  • What is the purpose of the video?

    -The purpose of the video is to guide professional artists and designers through the process of installing Invoke AI on Windows, a generative AI tool that combines the power of stable diffusion with professional workflows and a clean, intuitive user interface.

  • What are the technical requirements to run Invoke AI on Windows?

    -To run Invoke AI on Windows, you need an Nvidia-based graphics card with at least 4 GB of VRAM (8 GB for the latest version of stable diffusion called sdxl), at least 12 GB of RAM, at least 18 GB of free hard drive space, and the correct version of Python (3.10 or 3.11) installed on your system.

  • Is Invoke AI compatible with AMD graphics cards on Windows?

    -No, Invoke AI is not compatible with AMD graphics cards on Windows. However, it can be run on Linux with an AMD card.

  • What is the process to check the installed version of Python?

    -To check the installed version of Python, open the command prompt by typing 'CMD' in the start menu, then type 'python --version' to see the installed version.

  • How can one obtain a huggingface token for downloading models?

    -To obtain a huggingface token, one needs to visit huggingface.co/settings/tokens, sign in or create an account, and then generate a new token which can be copied to the clipboard.

  • What does Invoke AI offer for users who want to build their own generative AI tools?

    -Invoke AI offers an automated installer, support for adding existing stable diffusion models, and a user-friendly interface to play around with the program's features, making it easier for users to build their own generative AI tools.

  • How does the installation process of Invoke AI work?

    -The installation process involves downloading the installer from the Invoke AI GitHub page, extracting the files, running the install.bat file, and following the on-screen instructions to configure the installation, including selecting the GPU type, choosing the installation location, and setting up the base models.

  • What is the role of the Microsoft Visual C++ redistributable in the Invoke AI installation?

    -The Microsoft Visual C++ redistributable from 2015 to 2022 is a prerequisite for running Invoke AI. If it is not already installed on the system, users are required to download and install it as part of the Invoke AI setup process.

  • How can users add their existing stable diffusion models to Invoke AI?

    -Users can add their existing stable diffusion models to Invoke AI by going to the model manager within the program, selecting 'import models', and providing the path to where the models are stored on their system.

  • What is the process to run Invoke AI after installation?

    -After installation, users can run Invoke AI by navigating to the installation folder and double-clicking on the 'invoke.bat' file. They can then select the option to run the browser-based interface, and Invoke AI will launch in the browser.

  • How can users customize their Invoke AI experience?

    -Users can customize their Invoke AI experience by choosing which models to install during setup, selecting the installation location to avoid using the main drive, and configuring settings such as the output folder for images and accepting the creative ML responsible AI license.

Outlines

00:00

๐Ÿš€ Introduction to Invoke AI for Creative Professionals

The video script begins with an introduction aimed at professional artists and designers interested in using generative AI tools like Stable Diffusion. The host offers guidance on installing Invoke AI, a program that integrates Stable Diffusion with professional workflows and a user-friendly interface. The episode is sponsored by Everly Heights Patreon, which provides access to custom tools and models. The host also discusses the technical requirements for running Invoke AI on Windows, which include an Nvidia graphics card with sufficient VRAM, a specific amount of RAM, free hard drive space, and the correct version of Python. The process of installing Invoke AI, including downloading the installer from GitHub, extracting files, and setting up the program, is detailed step by step.

05:05

๐Ÿ“˜ Installing Invoke AI and Setting Up the Environment

The second paragraph continues the installation process of Invoke AI. It covers the steps to add Invoke AI to the Windows registry, selecting the installation location to avoidๅ ็”จ the main drive, and choosing the type of GPU, specifically noting the requirement of an Nvidia GPU that supports CUDA. The host opts for a manual configuration to sync with an existing Auto1111 setup. The process involves downloading models from Huggingface, which requires a token, and selecting the desired components and models for installation. The host also discusses the importance of saving images to a separate folder and accepting the Creative ML Responsible AI license before proceeding with the installation.

10:08

๐Ÿ–ฅ๏ธ Running Invoke AI and Testing with Models

The final paragraph of the script describes how to run Invoke AI after installation. The host explains how to launch the browser-based interface and add previously installed models into Invoke AI's model manager. It also demonstrates how to use the tool with a creative prompt, 'Superman drinking beer,' using a Stable Diffusion 1.5 model. The host shows the process of generating images from the prompt and expresses satisfaction with the results. The video concludes with an invitation for viewers to request in-depth tutorials on using Invoke AI for creative projects and to follow the development of the Everly Heights universe. Links to the founder's interview and Patreon support are provided, encouraging viewers to join the community and gain access to custom models and production diaries.

Mindmap

Keywords

๐Ÿ’กInvoke AI

Invoke AI is a generative AI art tool designed for professional artists and designers. It combines the capabilities of stable diffusion with professional workflows and a user-friendly interface. In the video, the host demonstrates how to install Invoke AI on a Windows system, highlighting its integration with stable diffusion models to create artwork efficiently.

๐Ÿ’กStable Diffusion

Stable Diffusion is a type of generative AI model used for creating images from textual descriptions. It's a core component of Invoke AI, allowing users to generate professional-grade artwork with minimal effort. The video mentions different versions of stable diffusion, including sdxl, which is the latest and most powerful version at the time of the video.

๐Ÿ’กNvidia Graphics Card

An Nvidia graphics card is a specific type of hardware required to run Invoke AI, particularly for handling the graphical processing demands of AI image generation. The video specifies that a card with at least 4 GB of VRAM is needed, or 8 GB for the latest stable diffusion models. This is a technical requirement that ensures the software can function correctly.

๐Ÿ’กPython

Python is a programming language that is essential for running Invoke AI. The video mentions the need for at least Python version 3.10 or 3.11, which can be checked and installed if not already present on the user's system. Python is used to execute the scripts that power the AI's functionality within Invoke AI.

๐Ÿ’กMicrosoft Visual C++ Redistributable

The Microsoft Visual C++ Redistributable is a set of runtime components that are required for some programs, including Invoke AI, to run properly on Windows. The video instructs viewers to install this if they don't already have it, as it's a prerequisite for the proper functioning of the AI tool.

๐Ÿ’กInstallation Process

The installation process for Invoke AI involves several steps, including downloading the installer from GitHub, extracting files, and running an installation script. The video provides a detailed walkthrough of this process, emphasizing the need for patience due to the time it takes for the installation to complete.

๐Ÿ’กHuggingface Token

A Huggingface token is used to access models from the Huggingface platform, which can be integrated into Invoke AI. The video shows how to obtain a token from Huggingface's website and use it during the installation process to download additional models.

๐Ÿ’กModel Manager

The Model Manager is a feature within Invoke AI that allows users to import and manage their stable diffusion models. The video demonstrates how to use the Model Manager to add existing models to Invoke AI, which is crucial for customizing the tool to the user's specific needs.

๐Ÿ’กCreative ML Responsible AI License

The Creative ML Responsible AI License is a type of license agreement that users must accept during the installation process. It outlines the terms of use for the AI models and ensures that users are aware of their responsibilities when utilizing the technology.

๐Ÿ’กEverly Heights

Everly Heights is a project mentioned in the video that utilizes Invoke AI and stable diffusion models to build a universe for a sitcom called 'Very Special.' The video is part of a series that supports this project, and the host encourages viewers to follow the progress on various social media platforms.

๐Ÿ’กAuto 1111 and Comfy UI

Auto 1111 and Comfy UI are other AI programs mentioned in the video that the host uses in conjunction with Invoke AI. They are part of the host's workflow for creating content for the Everly Heights universe, and the video suggests that viewers can integrate these tools for a more comprehensive creative process.

Highlights

The tutorial provides a step-by-step guide on installing Invoke AI on Windows for professional artists and designers.

Invoke AI combines the power of stable diffusion with professional workflows and a clean, intuitive UI.

Technical requirements for running Invoke on Windows include an Nvidia-based graphics card with at least 4 GB of VRAM.

At least 12 GB of RAM and 18 GB of free hard drive space are recommended for optimal performance.

Python version 3.10 or 3.11 is required, with instructions provided to check and update if necessary.

The Microsoft Visual C++ redistributable from 2015 to 2022 needs to be installed if not already present.

Invoke AI's GitHub page offers an automated installer for ease of setup.

The installation process includes configuring the GPU type and selecting an installation location.

Users can choose between an automatic configuration with base models or a manual configuration.

Huggingface token is required if downloading models from Huggingface, with instructions on how to obtain one.

The program allows for customizing the image output folder to integrate with existing databases or workflows.

Creative ML Responsible AI license acceptance is part of the installation process.

Models can be added to Invoke AI through the model manager after installation.

Invoke AI offers a browser-based interface for ease of use.

The tutorial demonstrates creating an image of Superman drinking beer using Invoke AI with a 1.5 model.

The video provides an example of the image generation process, showcasing the animation of images being created.

The founder of Invoke AI, Kent Kiry, discusses the tool's capabilities in an interview available on YouTube and as a podcast episode.

Support for the project is available through Patreon, where backers gain access to custom tools and stable diffusion models.