invoke-ai/InvokeAI v2.1.2 on GitHub

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

update mac instructions to use invokeai for env name by @willwillems in #1030
Update .gitignore by @blessedcoolant in #1040
reintroduce fix for m1 from #579 missing after merge by @skurovec in #1056
Update Stable_Diffusion_AI_Notebook.ipynb (Take 2) by @ChloeL19 in #1060
Print out the device type which is used by @manzke in #1073
Hires Addition by @hipsterusername in #1063
fix for "1 leaked semaphore objects to clean up at shutdown" on M1 by @skurovec in #1081
Forward dream.py to invoke.py using the same interpreter, add deprecation warning by @db3000 in #1077
fix noisy images at high step counts by @lstein in #1086
Generalize facetool strength argument by @db3000 in #1078
Enable fast switching among models at the invoke> command line by @lstein in #1066
Fix Typo, committed changing ldm environment to invokeai by @jdries3 in #1095
Update generate.py by @unreleased in #1109
Update 'ldm' env to 'invokeai' in troubleshooting steps by @19wolf in #1125
Fixed documentation typos and resolved merge conflicts by @rupeshs in #1123
Fix broken doc links, fix malaprop in the project subtitle by @majick in #1131
Only output facetool parameters if enhancing faces by @db3000 in #1119
Update gitignore to ignore codeformer weights at new location by @spezialspezial in #1136
fix links to point to invoke-ai.github.io #1117 by @mauwii in #1143
Rework-mkdocs by @mauwii in #1144
add option to CLI and pngwriter that allows user to set PNG compression level by @lstein in #1127
Fix img2img DDIM index out of bound by @wfng92 in #1137
Fix gh actions by @mauwii in #1128
update mac instructions to use invokeai for env name by @willwillems in #1030
Update .gitignore by @blessedcoolant in #1040
reintroduce fix for m1 from #579 missing after merge by @skurovec in #1056
Update Stable_Diffusion_AI_Notebook.ipynb (Take 2) by @ChloeL19 in #1060
Print out the device type which is used by @manzke in #1073
Hires Addition by @hipsterusername in #1063
fix for "1 leaked semaphore objects to clean up at shutdown" on M1 by @skurovec in #1081
Forward dream.py to invoke.py using the same interpreter, add deprecation warning by @db3000 in #1077
fix noisy images at high step counts by @lstein in #1086
Generalize facetool strength argument by @db3000 in #1078
Enable fast switching among models at the invoke> command line by @lstein in #1066
Fix Typo, committed changing ldm environment to invokeai by @jdries3 in #1095
Fixed documentation typos and resolved merge conflicts by @rupeshs in #1123
Only output facetool parameters if enhancing faces by @db3000 in #1119
add option to CLI and pngwriter that allows user to set PNG compression level by @lstein in #1127
Fix img2img DDIM index out of bound by @wfng92 in #1137
Add text prompt to inpaint mask support by @lstein in #1133
Respect http[s] protocol when making socket.io middleware by @damian0815 in #976
WebUI: Adds Codeformer support by @psychedelicious in #1151
Skips normalizing prompts for web UI metadata by @psychedelicious in #1165
Add Asymmetric Tiling by @carson-katri in #1132
Web UI: Increases max CFG Scale to 200 by @psychedelicious in #1172
Corrects color channels in face restoration; Fixes #1167 by @psychedelicious in #1175
Flips channels using array slicing instead of using OpenCV by @psychedelicious in #1178
Fix typo in docs: s/Formally/Formerly by @noodlebox in #1176
fix clipseg loading problems by @lstein in #1177
Correct color channels in upscale using array slicing by @wfng92 in #1181
Web UI: Filters existing images when adding new images; Fixes #1085 by @psychedelicious in #1171
fix a number of bugs in textual inversion by @lstein in #1190
Improve !fetch, add !replay command by @ArDiouscuros in #882
Fix generation of image with s>1000 by @holstvoogd in #951
Web UI: Gallery improvements by @psychedelicious in #1198
Update CLI.md by @krummrey in #1211
outcropping improvements by @lstein in #1207
add support for loading VAE autoencoders by @lstein in #1216
remove duplicate fix_func for MPS by @wfng92 in #1210
Metadata storage and retrieval fixes by @lstein in #1204
nix: add shell.nix file by @Cloudef in #1170
Web UI: Changes vite dist asset paths to relative by @psychedelicious in #1185
Web UI: Removes isDisabled from PromptInput by @psychedelicious in #1187
Allow user to generate images with initial noise as on M1 / mps system by @ArDiouscuros in #981
feat: adding filename format template by @plucked in #968
Web UI: Fixes broken bundle by @psychedelicious in #1242
Support runwayML custom inpainting model by @lstein in #1243
Update IMG2IMG.md by @talitore in #1262
New dockerfile - including a build- and a run- script as well as a GH-Action by @mauwii in #1233
cut over from karras to model noise schedule for higher steps by @lstein in #1222
Prompt tweaks by @lstein in #1268
Outpainting implementation by @Kyle0654 in #1251
fixing aspect ratio on hires by @tjennings in #1249
Fix-build-container-action by @mauwii in #1274
handle all unicode characters by @damian0815 in #1276
adds models.user.yml to .gitignore by @JakeHL in #1281
remove debug branch, set fail-fast to false by @mauwii in #1284
Protect-secrets-on-pr by @mauwii in #1285
Web UI: Adds initial inpainting implementation by @psychedelicious in #1225
fix environment-mac.yml - tested on x64 and arm64 by @mauwii in #1289
Use proper authentication to download model by @mauwii in #1287
Prevent indexing error for mode RGB by @spezialspezial in #1294
Integrate sd-v1-5 model into test matrix (easily expandable), remove unecesarry caches by @mauwii in #1293
add --no-interactive to preload_models step by @mauwii in #1302
1-click installer and updater. Uses micromamba to install git and conda into a contained environment (if necessary) before running the normal installation script by @cmdr2 in #1253
preload_models.py script downloads the weight files by @lstein in #1290

New Contributors

@willwillems made their first contribution in #1030
@ChloeL19 made their first contribution in #1060
@manzke made their first contribution in #1073
@rupeshs made their first contribution in #1123
@wfng92 made their first contribution in #1137
@carson-katri made their first contribution in #1132
@noodlebox made their first contribution in #1176
@krummrey made their first contribution in #1211
@Cloudef made their first contribution in #1170
@talitore made their first contribution in #1262
@tjennings made their first contribution in #1249
@JakeHL made their first contribution in #1281
@cmdr2 made their first contribution in #1253

Full Changelog: v2.0.1...2.1.0-rc1

invoke-ai/InvokeAI v2.1.2 InvokeAI Version 2.1 - A Stable Diffusion Toolkit on GitHub