FoooXus uses Fooocus Gradio API to extend Fooocus functionnalities :
- Manage your Styles, Models and LoRAs with preset prompts
- Simply add or remove presets in config.json file
- View all your images presets
- Launch image creation works to a queue that is processed in background
- Create generation batches with variations on some parameters (not ready yet)
You need a running Fooocus instance to use FoooXus. You must install FoooXus on the same device than Fooocus.
NOT AVAILABLE ON FOOOCUS CURRENT RELEASE See this document to use foooxus.exe executable
Natively, FoooXus is a python app server that launches a web UI and connect to Fooocus via API Install FoooXus following these standard steps.
# Get the sources
git clone https://github.com/toutjavascript/FoooXus-Fooocus-Extender.git
cd FoooXus-Fooocus-Extender
# Activate the virtual python environment
python -m venv venv
venv\Scripts\activate
# Install requirements
pip install -r requirements.txt
# First run of the FoooXus app to initialize
python foooxus.py
Terminal will show this kind of message
Fill in the fields in config.json file
FoooXus is now ready to start with a new call to
python foooxus.py
On Windows, you can double-click foooxus.bat
Or if you prefer
# Check you are in your (venv) virtual environment
# if not, activate it
venv\Scripts\activate
# Run the FoooXus app
python foooxus.py
Terminal shows now versions and much more informations to understand what happens
Open your browser on http://127.0.0.1:7878 to view FoooXus UI
To update FoooXus from this github repository, open a terminal in your FoooXus-Fooocus-Extender
git pull
FoooXus can now force Performance Settings when those models are used. New button to regenerate Models illustrations are displayed
- Fooocus could be updated at every moment. It could broke the API.
- FoooXus version is tested with V2.4.1 Fooocus Release.
V1.0.0 Update to V2.5.5 Fooocus Release
V0.9.7 Update to V2.4.1 Fooocus Release
V0.9.6 New optional loras-directory parameter on config.json To fix issue toutjavascript#7 where Loras are not found, you can add a new line in config.json. For example: "loras-directory": "D:\datas\StabilityMatrix\Data\Models\Lora",
V0.9.5 Force optimized Performance Settings when Turbo or Lightning models are used
V0.9.1 Adapt to Gradio API to Fooocus 2.3.0 Auto clean /outputs/tmp/ folder at launch
V0.9.0 Adapt to Gradio API to Fooocus 2.2.1
V0.8.3 Fix issue on macOS toutjavascript#5
V0.8.2 Add a button to clear queue
V0.8.1 Improve the error messages
V0.8 New standalone fooocus.exe
- new config.json options
- download the fooocus.exe executable file to avoid tedious python, git, pip installations
V0.6 Updates on install process
- If you have issues, please tell me, with a copy of a terminal
V0.5 Very First release
- Please, tell me if it is broken somewhere
- Please, share your ideas by opening new discussions
Get notified when new image is generated
View history
Python terminal shows all generations
FoooXus should work on Linux too