I primarily wrote this app because I wanted to generate a bunch of images using Stability AI without having to wait for the result. Just pass in one prompt after the other and have the app call the API and store the result.
Right now only Stability AI API is supported. I plan to add support for more models soon.
If you are looking to run models locally, then you should take a look at comfy ui
-
Clone the repository
-
Install requirements
pip install -r requirements.txt
-
Setting up API credential
a. you can either set the environment variable STABILITY_API_KEY
Windows
set STABILITY_API_KEY=add_key_here
Linux and macOS
export STABILITY_API_KEY=add_key_here
b. Or create a streamlit secrets files
mkdir .streamlit
touch .streamlit/secrets.toml
edit the secrets.toml file
stability_api_key = 'add_key_here' 4.Launch the app
streamlit run main.py
The app will be accessible at : localhost:8501.
- Build the image
docker build -t sigi .
Alternatively
You can download the image from docker hub
docker pull lecaptain/sigi:latest
- Run : provide your stability api key
docker run -p 8501:8501 -v sigi-data:/usr/src/app/ -e STABILITY_API_KEY=add_key_here sigi
The app will be accessible at : localhost:8501.