Essential AUTOMATIC1111 Extensions

Stable Diffusion is a powerful AI image generator. Its community-developed extensions make it stand out, enhancing its functionality and ease of use. Extensions shape our workflow and make Stable Diffusion even more powerful.

Software

We will use AUTOMATIC1111 Stable Diffusion GUI. You can use this GUI on WindowsMac, or Google Colab.

Check out the Quick Start Guide if you are new to Stable Diffusion.

Installing an extension on Windows or Mac

To install an extension in AUTOMATIC1111 Stable Diffusion WebUI:

  1. Start AUTOMATIC1111 Web-UI normally.

2. Navigate to the Extension Page.

3. Click the Install from URL tab.

4. Enter the extension’s URL in the URL for extension’s git repository field.

5. Click the Install button.

6. Wait for the confirmation message that the installation is complete.

7. Restart AUTOMATIC1111.

Updating an extension

Extensions need to be updated regularly to get bug fixes or new functionality. To update an extension:

  1. Go to the Extensions page.
  2. Click Installed tab.
  3. Click Check for updates.

If an update to an extension is available, you will see a new commits checkbox in the Update column.

Leave the checkbox checked for the extensions you wish to update.

Click Apply and Restart UI to update the extensions.

You may need to restart the AUTOMATIC1111 server manually for it to work.

Installing extensions in Google Colab

To install an extension in our Colab notebook:

  • Select the extension if it is available in the Extension section.
  • If unavailable, put the extension’s URL in the Extension_from_URL field.

Must-have extensions for AUTOMATIC1111

Here are the extensions I found essential for using Stable Diffusion daily.

ControlNet

https://github.com/Mikubill/sd-webui-controlnet

ControlNet has changed Stable Diffusion forever. It lets you copy human poses, color, and content of a reference image, to name a few. Mastering ControlNet is a logical next step after mastering prompt building.

Infinite Image Browser

  • Extension’s URL:
https://github.com/zanllp/sd-webui-infinite-image-browsing

You have many images. Want to find the older ones? Instead of using File Explorer (Windows) or Finder (Mac), you can view and search them directly in AUTOMATIC1111 using the Infinite Image Browser.

Plus, you can search for images based on prompts and models.

The Image Browser is especially useful when accessing A1111 from another machine, where browsing images is not easy.

Aspect Ratio selector

Extension’s URL

https://github.com/alemelis/sd-webui-ar

This Aspect Ratio Selector extension is for you if you are tired of remembering the pixel numbers for various aspect ratios. This simple extension populates the correct image size with a single mouse click.

A very nice feature is defining presets. Edit the file resolutions.txt in the extension’s folder (stable-diffusion-webuiextensionssd-webui-ar). Below are the presets I use. It is convenient to use these presets to switch between image sizes of SD 1.5 and SDXL.

SD1:1, 512, 512 # 1:1 square
SD3:2, 768, 512 # 3:2 landscape
XL1:1, 1024, 1024
XL5:4, 1152, 896
XL3:2, 1216, 832
XL16:9, 1344, 768
XL21:9, 1536, 640

CLIP Interrogator

  • Extension’s URL
https://github.com/pharmapsychotic/clip-interrogator-ext

The CLIP interrogator lets you query the prompt of an image. Although you can use the Interrogate CLIP function on the Img2img page to find a prompt of an image, this extension gives you more options. You can specify which language model to use and analyze an image with different databases.

Ultimate SD Upscale

https://github.com/Coyote-A/ultimate-upscale-for-automatic1111

The best resolution for Stable Diffusion v1 is 512 pixels on one side. Ultimate SD Upscale allows you to enlarge images with ControlNet and tile by tile. It is helpful if you don’t have a GPU card with a large VRAM.

Civitai Helper

Extension’s URL:

https://github.com/butaixianran/Stable-Diffusion-Webui-Civitai-Helper

If you use Stable Diffusion, you probably have downloaded a model from Civitai. But you must ensure putting the checkpoint, LoRA, and textual inversion models in the right folders. Civitai Helper lets you download models from Civitai right in the AUTOMATIC1111 GUI.

As a bonus, the cover image of the models will be downloaded. Use the Scan button to download the model images of existing models.

After Detailer

https://github.com/Bing-su/adetailer

After Detailer automatically fix faces and hands. It greatly increases the chance of getting a good image.

Regional Prompter

https://github.com/hako-mikan/sd-webui-regional-prompter

Regional Prompter allows you to split an image into sections and set unique prompts for each. This tool offers great flexibility: you can position objects accurately and choose specific colors for parts of the image without altering the rest.

Nice-to-have extensions

These extensions are pretty useful, but I won’t say I cannot live without them.

OpenPose Editor

Extension’s URL:

https://github.com/fkunn1326/openpose-editor

OpenPose Editor allows you to edit the control image from ControlNet‘s Openpose preprocessor. Now you can adjust the pose precisely when using ControlNet.

Style Selector for SDXL 1.0

https://github.com/ahgsql/StyleSelectorXL

Style Selector for SDXL conveniently adds preset keywords to prompts and negative prompts to achieve certain styles. The styles available are the same as those in DreamStudio, the official Stable Diffusion image generator.

Roop

https://github.com/s0md3v/sd-webui-roop

Roop is an easy-to-use extension to swap faces. You can use it with text-to-image or image-to-image.

Extensions for making videos

Here’s a list of AUTOMATIC1111 extensions you can use to make videos.

Deforum

https://github.com/deforum-art/sd-webui-deforum

Deforum is a one-of-a-kind video that you can make using Stable Diffusion. It looks psychedelic and is very fun to watch.

AnimateDiff

https://github.com/continue-revolution/sd-webui-animatediff

AnimateDiff lets you make short videos using prompts with any Stable Diffusion model.

text2video

https://github.com/kabachuha/sd-webui-text2video

Text2video is an extension for making videos using the ModelScope model.

Mov2mov

https://github.com/Scholar01/sd-webui-mov2mov

Mov2mov extension is a simple way to stylize a video using any Stable Diffusion model.

Temporal Kit

https://github.com/CiaraStrawberry/TemporalKit

Temporal Kit is a helper extension for using EbSynth, an app that lets you style videos frame by frame. Thanks to Stable Diffusion, you can now automatically paint keyframes, making EbSynth more popular than before.

It’s quite some work to use this extension, but it gives the best result.

aizmin: