Skip to content

Releases: invoke-ai/InvokeAI

InvokeAI Version 2.3.0

03 Feb 03:26
@lstein lstein
v2.3.0
f5d1fbd
This commit was created on GitHub.com and signed with GitHub’s verified signature. The key has expired.
GPG key ID: 4AEE18F83AFDEB23
Expired
Learn about vigilant mode.
Compare
Choose a tag to compare
InvokeAI Version 2.3.0

We are pleased to announce a features and performance update to InvokeAI with the release of version 2.3.0.

What's New in 2.3.0

There are multiple internal and external changes in this version of InvokeAI which greatly enhance the developer and user experiences respectively.

Migration to Stable Diffusion diffusers models

Previous versions of InvokeAI supported the original model file format introduced with Stable Diffusion 1.4. In the original format, known variously as "checkpoint", or "legacy" format, there is a single large weights file ending with .ckpt or .safetensors. Though this format has served the community well, it has a number of disadvantages, including file size, slow loading times, and a variety of non-standard variants that require special-case code to handle. In addition, because checkpoint files are actually a bundle of multiple machine learning sub-models, it is hard to swap different sub-models in and out, or to share common sub-models. A new format, introduced by the StabilityAI company in collaboration with HuggingFace, is called diffusers and consists of a directory of individual models. The most immediate benefit of diffusers is that they load from disk very quickly. A longer term benefit is that in the near future diffusers models will be able to share common sub-models, dramatically reducing disk space when you have multiple fine-tune models derived from the same base.

When you perform a new install of version 2.3.0, you will be offered the option to install the diffusers versions of a number of popular SD models, including Stable Diffusion versions 1.5 and 2.1 (including the 768x768 pixel version of 2.1). These will act and work just like the checkpoint versions. Do not be concerned if you already have a lot of ".ckpt" or ".safetensors" models on disk! InvokeAI 2.3.0 can still load these and generate images from them without any extra intervention on your part.

To take advantage of the optimized loading times of diffusers models, InvokeAI offers options to convert legacy checkpoint models into optimized diffusers models. If you use the invokeai command line interface, the relevant commands are:

  • !convert_model -- Take the path to a local checkpoint file or a URL that is pointing to one, convert it into a diffusers model, and import it into InvokeAI's models registry file.
  • !optimize_model -- If you already have a checkpoint model in your InvokeAI models file, this command will accept its short name and convert it into a like-named diffusers model, optionally deleting the original checkpoint file.
  • !import_model -- Take the local path of either a checkpoint file or a diffusers model directory and import it into InvokeAI's registry file. You may also provide the ID of any diffusers model that has been published on the HuggingFace models repository and it will be downloaded and installed automatically.

The WebGUI offers similar functionality for model management.

For advanced users, new command-line options provide additional functionality. Launching invokeai with the argument --autoconvert <path to directory> takes the path to a directory of checkpoint files, automatically converts them into diffusers models and imports them. Each time the script is launched, the directory will be scanned for new checkpoint files to be loaded. Alternatively, the --ckpt_convert argument will cause any checkpoint or safetensors model that is already registered with InvokeAI to be converted into a diffusers model on the fly, allowing you to take advantage of future diffusers-only features without explicitly converting the model and saving it to disk.

Please see INSTALLING MODELS for more information on model management in both the command-line and Web interfaces.

Support for the XFormers Memory-Efficient Crossattention Package

On CUDA (Nvidia) systems, version 2.3.0 supports the XFormers library. Once installed, thexformers package dramatically reduces the memory footprint of loaded Stable Diffusion models files and modestly increases image generation speed. xformers will be installed and activated automatically if you specify a CUDA system at install time.

The caveat with using xformers is that it introduces slightly non-deterministic behavior, and images generated using the same seed and other settings will be subtly different between invocations. Generally the changes are unnoticeable unless you rapidly shift back and forth between images, but to disable xformers and restore fully deterministic behavior, you may launch InvokeAI using the --no-xformers option. This is most conveniently done by opening the file invokeai/invokeai.init with a text editor, and adding the line --no-xformers at the bottom.

A Negative Prompt Box in the WebUI

There is now a separate text input box for negative prompts in the WebUI. This is convenient for stashing frequently-used negative prompts ("mangled limbs, bad anatomy"). The [negative prompt] syntax continues to work in the main prompt box as well.

To see exactly how your prompts are being parsed, launch invokeai with the --log_tokenization option. The console window will then display the tokenization process for both positive and negative prompts.

Model Merging

Version 2.3.0 offers an intuitive user interface for merging up to three Stable Diffusion models using an intuitive user interface. Model merging allows you to mix the behavior of models to achieve very interesting effects. To use this, each of the models must already be imported into InvokeAI and saved in diffusers format, then launch the merger using a new menu item in the InvokeAI launcher script (invoke.sh, invoke.bat) or directly from the command line with invokeai-merge --gui. You will be prompted to select the models to merge, the proportions in which to mix them, and the mixing algorithm. The script will create a new merged diffusers model and import it into InvokeAI for your use.

See MODEL MERGING for more details.

Textual Inversion Training

Textual Inversion (TI) is a technique for training a Stable Diffusion model to emit a particular subject or style when triggered by a keyword phrase. You can perform TI training by placing a small number of images of the subject or style in a directory, and choosing a distinctive trigger phrase, such as "pointillist-style". After successful training, The subject or style will be activated by including <pointillist-style> in your prompt.

Previous versions of InvokeAI were able to perform TI, but it required using a command-line script with dozens of obscure command-line arguments. Version 2.3.0 features an intuitive TI frontend that will build a TI model on top of any diffusers model. To access training you can launch from a new item in the launcher script or from the command line using invokeai-ti --gui.

See TEXTUAL INVERSION for further details.

A New Installer Experience

The InvokeAI installer has been upgraded in order to provide a smoother and hopefully more glitch-free experience. In addition, InvokeAI is now packaged as a PyPi project, allowing developers and power-users to install InvokeAI with the command pip install InvokeAI --use-pep517. Please see Installation for details.

Developers should be aware that the pip installation procedure has been simplified and that the conda method is no longer supported at all. Accordingly, the environments_and_requirements directory has been deleted from the repository.

Installation

To install or upgrade to InvokeAI 2.3.0, please download the zip file below, unpack it, and then double-click to launch the script install.sh (Macintosh, Linux) or install.bat (Windows). Alternatively, you can open a command-line window and execute the installation script directly.

InvokeAI-installer-v2.3.0.zip

If you are upgrading from an earlier version of InvokeAI, all you have to do is to run the installer for your platform. When the installer asks you to confirm the location of the invokeai directory, type in the path to the directory you are already using, if not the same as the one selected automatically by the installer. When the installer asks you to confirm that you want to install into an existing directory, simply indicate "yes".

Developers and power users can upgrade to the current version by activating the InvokeAI environment and then using pip install --use-pep517 --upgrade InvokeAI . You may specify a particular version by adding the version number to the command, as in InvokeAI==2.3.1. You can see which versions are available by going to The PyPI InvokeAI Project Page

Command-line name changes

All of InvokeAI's functionality, including the WebUI, command-line interface, textual inversion training and model merging, can all be accessed from the invoke.sh and invoke.bat launcher scripts. The menu of options has been expanded to add the new functionality. For the convenience of developers and power users, we have normalized the names of the InvokeAI command-line scripts:

  • invokeai -- Command-line client
  • invokeai --web -- Web GUI
  • `invokeai-me...
Read more

Contributors

shapor, xrd, and 37 other contributors
Assets 3
Loading
84 Join discussion

InvokeAI Version 2.2.5 - A Stable Diffusion Toolkit

01 Jan 19:35
@lstein lstein
v2.2.5p2
11601d4
Compare
Choose a tag to compare
InvokeAI Version 2.2.5 - A Stable Diffusion Toolkit

We are pleased to announce a features and bugfix update to InvokeAI with the release of version 2.2.5.

What's New in 2.2.5

WebUI

  • The WebGUI now features a Model Manager that lets you load and edit models interatively. It also allows you to pick a folder to scan and import new .ckpt files @blessedcoolant
  • Add Unified Canvas Alternate UI Beta: We added a new alternative UI to the Unified Canvas that mimics traditional photo editing applications you might be familiar with. You can switch to this new UI in the Settings menu by activating the new toggle option. @blessedcoolant
  • Restore and Upscale hotkeys have been changed from ‘R’ and ‘U’ to ‘Shift+R’ and ‘Shift+U’ respectively. This was done to avoid accidental keystrokes triggering these operations. @blessedcoolant
  • Added Localization. Support has been added for Russian, Italian, Portuguese (Brazilian), German, Polish @blessedcoolant
    Translators:

If you are interested in translating InvokeAI to your language, please feel free to reach out to us on Discord.

CLI

  • Add the --karras_max option to the command line. @lstein
  • Add the –version option to get the version of the app. @lstein
  • Remove requirement for Hugging Face token, now that it is no longer rqeuired. @ebr

Docker

  • Optimize dockerfile. @mauwii
  • Allow usage of GPU’s in Docker. @xrd

Bug Fixes & Updates

  • Fix not being able to load the model while inpainting when using the free_gpu_mem option. @rmagur1203
  • Various installer improvements. @lstein
  • Fix segfault error on MacOS when using homebrew. @ebr
  • Fix a None type error when nsfw_checker was turned on. @limonspb
  • Fix the number of tokens to cap to 75 and handle blends accordingly. @damian0815
  • [CLI] Fix the time step not displaying correctly during img2img. @wfng92
  • [WebUI] Fix the initial theme setting not displaying correctly in the selector after reload. @kasbah
  • [WebUI] Fix of Hires Fix on Img2Img tab @hipsterusername
  • Fix embeddings not working correctly. @blessedcoolant
  • Fix an issue where the —config launch argument was not being recognized. @blessedcoolant
  • Retrieve threshold from an image even if it is 0. @JPPhoto
  • Add –root_dir as an alternate arg for –root during launch.
  • Relax HuggingFace login requirements during setup. @ebr
  • Fixed an issue where the --no-patchmatch would not work. @lstein
  • Fixed a crash in img2img @lstein
  • Documentation, updates, typos and fixes. @limonspb, @lstein, @hipsterusername, @mauwii

Developer

  • Add concurrency to Github actions. @mauwii
  • Github action to lint python files with pyflakes @keturn
  • Fix circular dependencies on the frontend @kasbah
  • Add Github action for linting the frontend. @kasbah
  • Fix all linting warnings on the frontend. @kasbah
  • Add auto formatting for the frontend. @kasbah

New Contributors

Full Changelog: v2.2.4...latest

Installation

To install InvokeAI 2.2.5 on a new system, please download the zip file below, unzip it, and run the script install.sh (Macintosh, Linux) or install.bat (Windows). A walkthrough can be found at Installation Overview .
InvokeAI-installer-v2.2.5p2-linux.zip
InvokeAI-installer-v2.2.5p2-mac.zip
InvokeAI-installer-v2.2.5p2-windows.zip

Upgrading

If you have InvokeAI 2.2.4 installed, you can upgrade it quickly using an update script. Download the zip file below, and unpack it. Place the file update.bat (Windows) or update.sh (Linux/Mac) into your invokeai folder, replacing the update script that was previously there. Then launch the new update script from the command line or by double-clicking.

InvokeAI-updater-v2.2.5p2.zip

Help

Please see the InvokeAI Issues Board or the InvokeAI Discord for assistance from the development team.

Contributors

shapor, xrd, and 18 other contributors
Assets 6
Loading
19 Join discussion

InvokeAI Version 2.2.4 - A Stable Diffusion Toolkit

11 Dec 05:39
@lstein lstein
v2.2.4
bd0c0d7
This commit was created on GitHub.com and signed with GitHub’s verified signature. The key has expired.
GPG key ID: 4AEE18F83AFDEB23
Expired
Learn about vigilant mode.
Compare
Choose a tag to compare
InvokeAI Version 2.2.4 - A Stable Diffusion Toolkit

With InvokeAI 2.2, this project now provides enthusiasts and professionals a robust workflow solution for creating AI-generated and human facilitated compositions. Additional enhancements have been made as well, improving safety, ease of use, and installation.

Optimized for efficiency, InvokeAI needs only ~3.5GB of VRAM to generate a 512x768 image (and less for smaller images), and is compatible with Windows/Linux/Mac (M1 & M2).

You can see the release video here, which introduces the main WebUI enhancement for version 2.2 - The Unified Canvas. This new workflow is the biggest enhancement added to the WebUI to date, and unlocks a stunning amount of potential for users to create and iterate on their creations. The following sections describe what's new for InvokeAI.

Version 2.2.4 is a bugfix release. The major user-visible change is that we have overhauled the installation experience to make it faster and more stable. Please see Installation Overview for instructions on using the new installer, and see the .zip files in the Assets section below for the installer for your preferred platform. Note that you will need to install Python 3.9 or 3.10 to use the new installation method.

The new installers are located here. They have been updated 13 December in order to prevent a segfault crash on certain Macintosh systems.

There are a number of installation-related changes that previous InvokeAI users should be aware of:

Everything now lives in the invokeai directory.

Previously there were two directories to worry about, the directory that contained the InvokeAI source code and the launcher scripts, and the invokeai directory that contained the models files, embeddings, configuration and outputs. With the 2.2.4 release, this dual system is done away with, and everything, including the invoke.bat and invoke.sh launcher scripts, now live in a directory named invokeai. By default this directory is located in your home directory (e.g. \Users\yourname on Windows), but you can select where it goes at install time.

InvokeAI-installer-2.2.4-p5-linux.zip
InvokeAI-installer-2.2.4-p5-mac.zip
InvokeAI-installer-2.2.4-p5-windows.zip

After installation, you can delete the install directory (the one that the zip file creates when it unpacks). Do not delete or move the invokeai directory!

The .invokeai initialization file has been renamed invokeai/invokeai.init

You can place frequently-used startup options in this file, such as the default number of steps or your preferred sampler. To keep everything in one place, this file has now been moved into the invokeai directory and is named invokeai.init.

To update from Version 2.2.3

The easiest route is to download and unpack one of the 2.2.4 installer files. When it asks you for the location of the invokeai runtime directory, respond with the path to the directory that contains your 2.2.3 invokeai. That is, if invokeai lives at C:\Users\fred\invokeai, then answer with C:\Users\fred and answer "Y" when asked if you want to reuse the directory.

The update.sh (update.bat) script that came with the 2.2.3 source installer does not know about the new directory layout and won't be fully functional.

To update to 2.2.5 (and beyond) there's now an update path.

As they become available, you can update to more recent versions of InvokeAI using an update.sh (update.bat) script located in the invokeai directory. Running it without any arguments will install the most recent version of InvokeAI. Alternatively, you can get set releases by running the update.sh script with an argument in the command shell. This syntax accepts the path to the desired release's zip file, which you can find by clicking on the green "Code" button on this repository's home page. Here are some examples:

# 2.2.4 release
update.sh https://github.com/invoke-ai/InvokeAI/archive/refs/tags/v2.2.4.zip

# 2.2.5 release  (don't try; it doesn't exist yet!)
update.sh https://github.com/invoke-ai/InvokeAI/archive/refs/tags/v2.2.5.zip

# current development version
update.sh https://github.com/invoke-ai/InvokeAI/archive/main.zip

# feature branch 3d-movies (don't try; it doesn't exist yet!)
update.sh https://github.com/invoke-ai/InvokeAI/archive/3d-movies.zip

Other 2.2.4 Improvements

New Contributors

Full Changelog: v2.2.3...v2.2.4

Contributors

lstein, damian0815, and 13 other contributors
Assets 5
Loading

InvokeAI 2.2.3

02 Dec 17:54
@lstein lstein
v2.2.3
b487fa4
Compare
Choose a tag to compare
InvokeAI 2.2.3

Note: This point release removes references to the binary installer from the installation guide. The binary installer is not stable at the current time. First time users are encouraged to use the "source" installer as described in Installing InvokeAI with the Source Installer

With InvokeAI 2.2, this project now provides enthusiasts and professionals a robust workflow solution for creating AI-generated and human facilitated compositions. Additional enhancements have been made as well, improving safety, ease of use, and installation.

Optimized for efficiency, InvokeAI needs only ~3.5GB of VRAM to generate a 512x768 image (and less for smaller images), and is compatible with Windows/Linux/Mac (M1 & M2).

You can see the release video here, which introduces the main WebUI enhancement for version 2.2 - The Unified Canvas. This new workflow is the biggest enhancement added to the WebUI to date, and unlocks a stunning amount of potential for users to create and iterate on their creations. The following sections describe what's new for InvokeAI.

Update 1 December 2022 -

  • The Unified Canvas: The Web UI now features a fully fitted infinite canvas that is capable of outpainting, inpainting, img2img and txt2img so you can streamline and extend your creative workflow. The canvas was rewritten to improve performance greatly and bring support for a variety of features like Paint Brushing, Unlimited History, Real-Time Progress displays and more.

  • Embedding Management: Easily pull from the top embeddings on Huggingface directly within Invoke, using the embed token to generate the exact style you want. With the ability to use multiple embeds simultaneously, you can easily import and explore different styles within the same session!

  • Viewer: The Web UI now also features a Viewer that lets you inspect your invocations in greater detail. No more opening the images in your external file explorer, even with large upscaled images!

  • 1 Click Installer Launch: With our official 1-click installation launch, using our tool has never been easier. Our OS specific bundles (Mac M1/M2, Windows, and Linux) will get everything set up for you. Click and get going - It’s now simple to get started with InvokeAI. See Installation.

  • Model Safety: A checkpoint scanner (picklescan) has been added to the initialization process of new models, helping with security against malicious and evil pickles.

  • DPM++2 Experimental Samplers: New samplers have been added! Please note that these are experimental, and are subject to change in the future as we continue to enhance our backend system.

First-time Installation

For those installing InvokeAI for the first time, please use this recipe:
For automated installation, open up the "Assets" section below and download one of the InvokeAI-*.zip files. The instructions in the Installation section of the InvokeAI docs will provide you with a guide to which file to download and what to do with it when you get it.

For manual installation download one of the "Source Code" archive files located in the Assets below.
Unpack the file, and enter the InvokeAI directory that it creates. Alternatively, you may clone the source code repository using the command git clone http://github.com/invoke-ai/InvokeAI and follow the instructions in Manual Installation.

Upgrading

For those wishing to upgrade from an earlier version, please use this recipe:
Download one of the "Source Code" archive files located in the Assets below.
Unpack the file, and enter the InvokeAI directory that it creates.
Alternatively, if you have previously cloned the InvokeAI repository, you may update it by entering the InvokeAI directory and running the commands git checkout main, followed by git pull
Select the appropriate environment file for your operating system and GPU hardware. A number of files can be found in a new environments-and-requirements directory:

environment-lin-amd.yml # Linux with an AMD (ROCm) GPU
environment-lin-cuda.yml # Linux with an NVIDIA CUDA GPU
environment-mac.yml # Macintoshes with MPS acceleration
environment-win-cuda.yml # Windows with an NVIDA CUDA GPU

Important Step that developers tend to miss! Either copy this environment file to the root directory with the name environment.yml, or make a symbolic link from environment.yml to the selected enrivonment file:

Macintosh and Linux using a symbolic link:
ln -sf environments-and-requirements/environment-xxx-yyy.yml environment.yml

Replace xxx and yyy with the appropriate OS and GPU codes.

Windows:
copy environments-and-requirements\environment-win-cuda.yml environment.yml

When this is done, confirm that a file environment.yml has been created in the InvokeAI root directory and that it points to the correct file in the environments-and-requirements directory.
Now run the following commands in the InvokeAI directory.

conda env update
conda activate invokeai
python scripts/preload_models.py

Additional installation information, including recipes for installing without Conda, can be found in Manual Installation

Known Bugs

  1. If you use the binary installer, the autocomplete function will not work on the command line client due to limitations of the version of python that the installer uses. However, all other functions of the command line client, and all features of the web UI will function perfectly well.
  2. The PyPatchMatch module, which provides excellent outpainting and inpainting results, does not currently work on Macintoshes. It will work on Linux after a support library is added to the system. See Installing PyPatchMatch.
  3. InvokeAI 2.2.0 does not support the Stable Diffusion 2.0 model at the current time, but is expected to provide full support in the near future.
  4. The 1650 and 1660ti GPU cards only run in full-precision mode, which greatly limits the size of the models you can load and images you can generate with InvokeAI.

Contributing
Please see CONTRIBUTORS for a list of the many individuals who contributed to this project. Also many thanks to the dozens of patient testers who flushed out bugs in this release before it went live.
Anyone who wishes to contribute to this project, whether documentation, features, bug fixes, code
cleanup, testing, or code reviews, is very much encouraged to do so. If you are unfamiliar with how
to contribute to GitHub projects, here is a
Getting Started Guide. Unlike previous versions of InvokeAI we have now moved all development to the main branch, so please make your pull requests against this branch.

Support
For support, please use this repository's GitHub Issues tracking service. Live support is also available on the InvokeAI Discord server.

Assets 5
Loading

InvokeAI Version 2.2.2 - A Stable Diffusion Toolkit

02 Dec 14:13
@lstein lstein
v2.2.2
c7960f9
Compare
Choose a tag to compare
InvokeAI Version 2.2.2 - A Stable Diffusion Toolkit

Note: The binary installer is not ready for prime time. First time users are recommended to install via the "source" installer accessible through the links at the bottom of this page.

With InvokeAI 2.2, this project now provides enthusiasts and professionals a robust workflow solution for creating AI-generated and human facilitated compositions. Additional enhancements have been made as well, improving safety, ease of use, and installation.

Optimized for efficiency, InvokeAI needs only ~3.5GB of VRAM to generate a 512x768 image (and less for smaller images), and is compatible with Windows/Linux/Mac (M1 & M2).

You can see the release video here, which introduces the main WebUI enhancement for version 2.2 - The Unified Canvas. This new workflow is the biggest enhancement added to the WebUI to date, and unlocks a stunning amount of potential for users to create and iterate on their creations. The following sections describe what's new for InvokeAI.

Update 1 December 2022 -

  • The Unified Canvas: The Web UI now features a fully fitted infinite canvas that is capable of outpainting, inpainting, img2img and txt2img so you can streamline and extend your creative workflow. The canvas was rewritten to improve performance greatly and bring support for a variety of features like Paint Brushing, Unlimited History, Real-Time Progress displays and more.

  • Embedding Management: Easily pull from the top embeddings on Huggingface directly within Invoke, using the embed token to generate the exact style you want. With the ability to use multiple embeds simultaneously, you can easily import and explore different styles within the same session!

  • Viewer: The Web UI now also features a Viewer that lets you inspect your invocations in greater detail. No more opening the images in your external file explorer, even with large upscaled images!

  • 1 Click Installer Launch: With our official 1-click installation launch, using our tool has never been easier. Our OS specific bundles (Mac M1/M2, Windows, and Linux) will get everything set up for you. Click and get going - It’s now simple to get started with InvokeAI. See Installation.

  • Model Safety: A checkpoint scanner (picklescan) has been added to the initialization process of new models, helping with security against malicious and evil pickles.

  • DPM++2 Experimental Samplers: New samplers have been added! Please note that these are experimental, and are subject to change in the future as we continue to enhance our backend system.

First-time Installation

For those installing InvokeAI for the first time, please use this recipe:
For automated installation, open up the "Assets" section below and download one of the InvokeAI-*.zip files. The instructions in the Installation section of the InvokeAI docs will provide you with a guide to which file to download and what to do with it when you get it.

For manual installation download one of the "Source Code" archive files located in the Assets below.
Unpack the file, and enter the InvokeAI directory that it creates. Alternatively, you may clone the source code repository using the command git clone http://github.com/invoke-ai/InvokeAI and follow the instructions in Manual Installation.

Upgrading

For those wishing to upgrade from an earlier version, please use this recipe:
Download one of the "Source Code" archive files located in the Assets below.
Unpack the file, and enter the InvokeAI directory that it creates.
Alternatively, if you have previously cloned the InvokeAI repository, you may update it by entering the InvokeAI directory and running the commands git checkout main, followed by git pull
Select the appropriate environment file for your operating system and GPU hardware. A number of files can be found in a new environments-and-requirements directory:

environment-lin-amd.yml # Linux with an AMD (ROCm) GPU
environment-lin-cuda.yml # Linux with an NVIDIA CUDA GPU
environment-mac.yml # Macintoshes with MPS acceleration
environment-win-cuda.yml # Windows with an NVIDA CUDA GPU

Important Step that developers tend to miss! Either copy this environment file to the root directory with the name environment.yml, or make a symbolic link from environment.yml to the selected enrivonment file:

Macintosh and Linux using a symbolic link:
ln -sf environments-and-requirements/environment-xxx-yyy.yml environment.yml

Replace xxx and yyy with the appropriate OS and GPU codes.

Windows:
copy environments-and-requirements\environment-win-cuda.yml environment.yml

When this is done, confirm that a file environment.yml has been created in the InvokeAI root directory and that it points to the correct file in the environments-and-requirements directory.
Now run the following commands in the InvokeAI directory.

conda env update
conda activate invokeai
python scripts/preload_models.py

Additional installation information, including recipes for installing without Conda, can be found in Manual Installation

Known Bugs

  1. If you use the binary installer, the autocomplete function will not work on the command line client due to limitations of the version of python that the installer uses. However, all other functions of the command line client, and all features of the web UI will function perfectly well.
  2. The PyPatchMatch module, which provides excellent outpainting and inpainting results, does not currently work on Macintoshes. It will work on Linux after a support library is added to the system. See Installing PyPatchMatch.
  3. InvokeAI 2.2.0 does not support the Stable Diffusion 2.0 model at the current time, but is expected to provide full support in the near future.
  4. The 1650 and 1660ti GPU cards only run in full-precision mode, which greatly limits the size of the models you can load and images you can generate with InvokeAI.

Contributing
Please see CONTRIBUTORS for a list of the many individuals who contributed to this project. Also many thanks to the dozens of patient testers who flushed out bugs in this release before it went live.
Anyone who wishes to contribute to this project, whether documentation, features, bug fixes, code
cleanup, testing, or code reviews, is very much encouraged to do so. If you are unfamiliar with how
to contribute to GitHub projects, here is a
Getting Started Guide. Unlike previous versions of InvokeAI we have now moved all development to the main branch, so please make your pull requests against this branch.

Support
For support, please use this repository's GitHub Issues tracking service. Live support is also available on the InvokeAI Discord server.

Assets 5
Loading

InvokeAI Version 2.2.0

30 Nov 16:35
@lstein lstein
v2.2.0
a47ef3d
Compare
Choose a tag to compare
InvokeAI Version 2.2.0

With InvokeAI 2.2, this project now provides enthusiasts and professionals a robust workflow solution for creating AI-generated and human facilitated compositions. Additional enhancements have been made as well, improving safety, ease of use, and installation.

Optimized for efficiency, InvokeAI needs only ~3.5GB of VRAM to generate a 512x768 image (and less for smaller images), and is compatible with Windows/Linux/Mac (M1 & M2).

You can see the release video here, which introduces the main WebUI enhancement for version 2.2 - The Unified Canvas. This new workflow is the biggest enhancement added to the WebUI to date, and unlocks a stunning amount of potential for users to create and iterate on their creations. The following sections describe what's new for InvokeAI.

Update 1 December 2022 -

  • The Unified Canvas: The Web UI now features a fully fitted infinite canvas that is capable of outpainting, inpainting, img2img and txt2img so you can streamline and extend your creative workflow. The canvas was rewritten to improve performance greatly and bring support for a variety of features like Paint Brushing, Unlimited History, Real-Time Progress displays and more.

  • Embedding Management: Easily pull from the top embeddings on Huggingface directly within Invoke, using the embed token to generate the exact style you want. With the ability to use multiple embeds simultaneously, you can easily import and explore different styles within the same session!

  • Viewer: The Web UI now also features a Viewer that lets you inspect your invocations in greater detail. No more opening the images in your external file explorer, even with large upscaled images!

  • 1 Click Installer Launch: With our official 1-click installation launch, using our tool has never been easier. Our OS specific bundles (Mac M1/M2, Windows, and Linux) will get everything set up for you. Click and get going - It’s now simple to get started with InvokeAI. See Installation.

  • Model Safety: A checkpoint scanner (picklescan) has been added to the initialization process of new models, helping with security against malicious and evil pickles.

  • DPM++2 Experimental Samplers: New samplers have been added! Please note that these are experimental, and are subject to change in the future as we continue to enhance our backend system.

First-time Installation

For those installing InvokeAI for the first time, please use this recipe:
For automated installation, open up the "Assets" section below and download one of the InvokeAI-*.zip files. The instructions in the Installation section of the InvokeAI docs will provide you with a guide to which file to download and what to do with it when you get it.

For manual installation download one of the "Source Code" archive files located in the Assets below.
Unpack the file, and enter the InvokeAI directory that it creates. Alternatively, you may clone the source code repository using the command git clone http://github.com/invoke-ai/InvokeAI and follow the instructions in Manual Installation.

Upgrading

For those wishing to upgrade from an earlier version, please use this recipe:
Download one of the "Source Code" archive files located in the Assets below.
Unpack the file, and enter the InvokeAI directory that it creates.
Alternatively, if you have previously cloned the InvokeAI repository, you may update it by entering the InvokeAI directory and running the commands git checkout main, followed by git pull
Select the appropriate environment file for your operating system and GPU hardware. A number of files can be found in a new environments-and-requirements directory:

environment-lin-amd.yml # Linux with an AMD (ROCm) GPU
environment-lin-cuda.yml # Linux with an NVIDIA CUDA GPU
environment-mac.yml # Macintoshes with MPS acceleration
environment-win-cuda.yml # Windows with an NVIDA CUDA GPU

Important Step that developers tend to miss! Either copy this environment file to the root directory with the name environment.yml, or make a symbolic link from environment.yml to the selected enrivonment file:

Macintosh and Linux using a symbolic link:
ln -sf environments-and-requirements/environment-xxx-yyy.yml environment.yml

Replace xxx and yyy with the appropriate OS and GPU codes.

Windows:
copy environments-and-requirements\environment-win-cuda.yml environment.yml

When this is done, confirm that a file environment.yml has been created in the InvokeAI root directory and that it points to the correct file in the environments-and-requirements directory.
Now run the following commands in the InvokeAI directory.

conda env update
conda activate invokeai
python scripts/preload_models.py

Additional installation information, including recipes for installing without Conda, can be found in Manual Installation

Known Bugs

  1. If you use the binary installer, the autocomplete function will not work on the command line client due to limitations of the version of python that the installer uses. However, all other functions of the command line client, and all features of the web UI will function perfectly well.
  2. The PyPatchMatch module, which provides excellent outpainting and inpainting results, does not currently work on Macintoshes. It will work on Linux after a support library is added to the system. See Installing PyPatchMatch.
  3. InvokeAI 2.2.0 does not support the Stable Diffusion 2.0 model at the current time, but is expected to provide full support in the near future.
  4. The 1650 and 1660ti GPU cards only run in full-precision mode, which greatly limits the size of the models you can load and images you can generate with InvokeAI.

Contributing
Please see CONTRIBUTORS for a list of the many individuals who contributed to this project. Also many thanks to the dozens of patient testers who flushed out bugs in this release before it went live.
Anyone who wishes to contribute to this project, whether documentation, features, bug fixes, code
cleanup, testing, or code reviews, is very much encouraged to do so. If you are unfamiliar with how
to contribute to GitHub projects, here is a
Getting Started Guide. Unlike previous versions of InvokeAI we have now moved all development to the main branch, so please make your pull requests against this branch.

Support
For support, please use this repository's GitHub Issues tracking service. Live support is also available on the InvokeAI Discord server.

Assets 8
Loading

InvokeAI Version 2.1.3 - A Stable Diffusion Toolkit

09 Nov 19:00
@lstein lstein
v2.1.3p1
93cdb47
Compare
Choose a tag to compare
InvokeAI Version 2.1.3 - A Stable Diffusion Toolkit

Welcome! Click here to get the latest release!

Read below for the old 2.1.3 release.

Invoke AI 2.1.3

The invoke-ai team is excited to be able to share the release of InvokeAI 2.1 - A Stable Diffusion Toolkit, a project that aims to provide enthusiasts and professionals both a suite of robust image creation tools. Optimized for efficiency, InvokeAI needs only ~3.5GB of VRAM to generate a 512x768 image (and less for smaller images), and is compatible with Windows/Linux/Mac (M1 & M2).

InvokeAI was one of the earliest forks of the core CompVis repo (formerly lstein/stable-diffusion), and recently evolved into a full-fledged community driven and open source stable diffusion toolkit. Version 2.1 of the tool introduces multiple new features and performance enhancements.

This 14-minute YouTube video introduces you to some of the new features contained in this release. The following sections describe what's new in the Web interface (WebGUI) and the command-line interface (CLI).

Version 2.1.3 is primarily a bug fix release that improves the installation process and provides enhanced stability and usability.

Update 22 November - updated invokeAI-src-installer-mac.zip to correct an error downloading the micromamba distirbution.

New features

  • A choice of installer scripts that automate installation and configuration. See Installation.
  • A streamlined manual installation process that works for both Conda and PIP-only installs. See Manual Installation.
  • The ability to save frequently-used startup options (model to load, steps, sampler, etc) in a .invokeai file. See Client
  • Support for AMD GPU cards (non-CUDA) on Linux machines.
  • Multiple bugs and edge cases squashed.

Installation

For those installing InvokeAI for the first time, please use this recipe:

  1. For automated installation, open up the "Assets" section below and download one of the InvokeAI-*.zip files. The instructions in the Installation section of the [InvokeAI docs](https://invoke-ai.github.io/InvokeAI) will provide you with a guide to which file to download and what to do with it when you get it.
  2. For manual installation download one of the "Source Code" archive files located in the Assets below.
    • Unpack the file, and enter the InvokeAI directory that it creates.
    • Alternatively, you may clone the source code repository using the command git clone http://github.com/invoke-ai/InvokeAI
    • Follow the instructions in Manual Installation.

Upgrading

For those wishing to upgrade from an earlier version, please use this recipe:

  1. Download one of the "Source Code" archive files located in the Assets below.
  2. Unpack the file, and enter the InvokeAI directory that it creates.
  3. Alternatively, if you have previously cloned the InvokeAI repository, you may update it by entering the InvokeAI directory and running the commands git checkout main, followed by git pull
  4. Select the appropriate environment file for your operating system and GPU hardware. A number of files can be found in a new environments-and-requirements directory:
environment-lin-amd.yml     # Linux with an AMD (ROCm) GPU
environment-lin-cuda.yml    # Linux with an NVIDIA CUDA GPU
environment-mac.yml         # Macintoshes with MPS acceleration
environment-win-cuda.yml    # Windows with an NVIDA CUDA GPU
  1. Important Step that developers tend to miss! Either copy this environment file to the root directory with the name environment.yml, or make a symbolic link from environment.yml to the selected enrivonment file:

  • Macintosh and Linux using a symbolic link:

    ln -sf environments-and-requirements/environment-xxx-yyy.yml environment.yml
# Replace `xxx` and `yyy` with the appropriate OS and GPU codes.

  • Windows:

copy environments-and-requirements\environment-win-cuda.yml environment.yml

When this is done, confirm that a file environment.yml has been created in the InvokeAI root directory and that it points to the correct file in the environments-and-requirements directory.

  1. Now run the following commands in the InvokeAI directory.
conda update
conda activate invokeai
python scripts/preload_models.py

Additional installation information, including recipes for installing without Conda, can be found in Manual Installation

Contributing

Please see CONTRIBUTORS for a list of the many individuals who contributed to this project. Also many thanks to the dozens of patient testers who flushed out bugs in this release before it went live.

Anyone who wishes to contribute to this project, whether documentation, features, bug fixes, code
cleanup, testing, or code reviews, is very much encouraged to do so. If you are unfamiliar with how
to contribute to GitHub projects, here is a
Getting Started Guide.

The most important thing is to know about contributing code is to make your pull request against the "development" branch, and not against "main". This will help keep public breakage to a minimum and will allow you to propose more radical
changes.

Support

For support, please use this repository's GitHub Issues tracking service. Live support is also available on the InvokeAI Discord server.

Assets 8
Loading
1 Join discussion

InvokeAI Version 2.1 - A Stable Diffusion Toolkit

03 Nov 22:52
@lstein lstein
v2.1.2
8aa4a25
Compare
Choose a tag to compare
InvokeAI Version 2.1 - A Stable Diffusion Toolkit

The invoke-ai team is excited to be able to share the release of InvokeAI 2.1 - A Stable Diffusion Toolkit, a project that aims to provide enthusiasts and professionals both a suite of robust image creation tools. Optimized for efficiency, InvokeAI needs only ~3.5GB of VRAM to generate a 512x768 image (and less for smaller images), and is compatible with Windows/Linux/Mac (M1 & M2).

InvokeAI was one of the earliest forks of the core CompVis repo (formerly lstein/stable-diffusion), and recently evolved into a full-fledged community driven and open source stable diffusion toolkit. Version 2.1 of the tool introduces multiple new features and performance enhancements.

This 14-minute YouTube video introduces you to some of the new features contained in this release. The following sections describe what's new in the Web interface (WebGUI) and the command-line interface (CLI).

Major new features

  • Inpainting support in the WebGUI
  • Greatly improved navigation and user experience in the WebGUI
  • The prompt syntax has been enhanced with prompt weighting, cross-attention and prompt merging.
  • You can now load multiple models and switch among them quickly without leaving the CLI or WebGUI.
  • The installation process (via scripts/preload_models.py) now lets you select among several popular Stable Diffusion models and downloads and installs them on your behalf. Among other models, this script will install the current Stable Diffusion 1.5 model as well as a StabilityAI variable autoencoder (VAE) which improves face generation.
  • Tired of struggling with photoeditors to get the masked region of for inpainting just right? Let the AI make the mask for you using text masking. This feature allows you to specify the part of the image to paint over using just English-language phrases.
  • Tired of seeing the head of your subjects cropped off? Uncrop them in the CLI with the outcrop feature.
  • Tired of seeing your subject's bodies duplicated or mangled when generating larger-dimension images? Check out the --hires option in the CLI, or select the corresponding toggle in the WebGUI.
  • We now support textual inversion and fine-tune .bin styles and subjects from the Hugging Face archive of SD Concepts. Load the .bin file using the --embedding_path option. (The next version will support merging and loading of multiple simultaneous models).

Installation

To install InvokeAI from scratch, please see the Installation section of the InvokeAI docs.

Upgrading

For those wishing to upgrade from an earlier version, please use the following recipe from within the InvokeAI directory:

Mac users:

conda deactivate
git checkout main
git pull
rm -rf src
conda update -f environment-mac.yml
conda activate invokeai
python scripts/preload_models.py

Windows users:

conda deactivate
git checkout main
git pull
rmdir src /s
conda update
conda activate invokeai
python scripts\preload_models.py

Linux Users

conda deactivate
git checkout main
git pull
rm -rf src
conda update
conda activate invokeai
python scripts/preload_models.py

Contributing

Please see CONTRIBUTORS for a list of the many individuals who contributed to this project. Also many thanks to the dozens of patient testers who flushed out bugs in this release before it went live.

Anyone who wishes to contribute to this project, whether documentation, features, bug fixes, code
cleanup, testing, or code reviews, is very much encouraged to do so. If you are unfamiliar with how
to contribute to GitHub projects, here is a
Getting Started Guide.

The most important thing is to know about contributing code is to make your pull request against the "development" branch, and not against "main". This will help keep public breakage to a minimum and will allow you to propose more radical
changes.

Support

For support, please use this repository's GitHub Issues tracking service. Live support is also available on the InvokeAI Discord server.

Full change log since 2.0.2

Read more

Contributors

holstvoogd, tjennings, and 28 other contributors
Assets 2
Loading
0 Join discussion

InvokeAI Version 2.0.2 - A Stable Diffusion Toolkit

18 Oct 20:35
@lstein lstein
v2.0.2
90d37ea
Compare
Choose a tag to compare
InvokeAI Version 2.0.2 - A Stable Diffusion Toolkit

The invoke-ai team is excited to be able to share the release of InvokeAI 2.0 - A Stable Diffusion Toolkit, a project that aims to provide enthusiasts and professionals both a suite of robust image creation tools. Optimized for efficiency, InvokeAI needs only ~3.5GB of VRAM to generate a 512x768 image (and less for smaller images), and is compatible with Windows/Linux/Mac (M1 & M2).

InvokeAI was one of the earliest forks of the core CompVis repo (formerly lstein/stable-diffusion), and recently evolved into a full-fledged community driven and open source stable diffusion toolkit named InvokeAI. Version 2.0.0 of the tool introduces an entirely new WebUI Front-end with a Desktop mode, and an optimized back-end server that can be interacted with via CLI or extended with your own fork.

Release 2.0.2 updates three Python dependencies that were recently reported to have critical security holes, and enhances documentation. Otherwise, the feature set is identical to 2.0.1.

This version of the app improves in-app workflows leveraging GFPGAN and Codeformer for face restoration, and RealESRGAN upscaling - Additionally, the CLI also supports a large variety of features:

  • Inpainting
  • Outpainting
  • Negative Prompts (prompt unconditioning)
  • Fast online model switching
  • Textual Inversion
  • Improved Quality for Hi-Resolution Images (Embiggen, Hi-res Fixes, etc.)
  • And more...

Future updates planned included UI driven outpainting/inpainting, robust Cross Attention support, and an advanced node workflow for automating and sharing your workflows with the community.

What's Changed

New Contributors

Full Changelog: v2.0.0...v2.0.1

What's Changed

New Contributors

Full Changelog: v2.0.1...v2.0.2

Contributors

lstein, majick, and 19 other contributors
Assets 2
Loading
0 Join discussion

InvokeAI Version 2.0.1 - A Stable Diffusion Toolkit

14 Oct 20:29
@lstein lstein
v2.0.1
0a7005f
Compare
Choose a tag to compare
InvokeAI Version 2.0.1 - A Stable Diffusion Toolkit

The invoke-ai team is excited to be able to share the release of InvokeAI 2.0 - A Stable Diffusion Toolkit, a project that aims to provide enthusiasts and professionals both a suite of robust image creation tools. Optimized for efficiency, InvokeAI needs only ~3.5GB of VRAM to generate a 512x768 image (and less for smaller images), and is compatible with Windows/Linux/Mac (M1 & M2).

InvokeAI was one of the earliest forks of the core CompVis repo (formerly lstein/stable-diffusion), and recently evolved into a full-fledged community driven and open source stable diffusion toolkit named InvokeAI. Version 2.0.0 of the tool introduces an entirely new WebUI Front-end with a Desktop mode, and an optimized back-end server that can be interacted with via CLI or extended with your own fork.

Release 2.0.1 corrects an error that was causing the k* samplers to produce noisy images at high step counts. Otherwise the feature set is the same as 2.0.0.

This version of the app improves in-app workflows leveraging GFPGAN and Codeformer for face restoration, and RealESRGAN upscaling - Additionally, the CLI also supports a large variety of features:

  • Inpainting
  • Outpainting
  • Negative Prompts (prompt unconditioning)
  • Fast online model switching
  • Textual Inversion
  • Improved Quality for Hi-Resolution Images (Embiggen, Hi-res Fixes, etc.)
  • And more...

Future updates planned included UI driven outpainting/inpainting, robust Cross Attention support, and an advanced node workflow for automating and sharing your workflows with the community.

What's Changed

New Contributors

Full Changelog: v2.0.0...v2.0.1

Contributors

Miserlou, tildebyte, and 5 other contributors
Assets 2
Loading