Running Locally

First, download the pre-trained weights:

cog run script/download-weights

If you get an error, you may need to chmod 777 script/download-weights

Build the docker image:

cog build -t sdxl-turbo-cog

Run the docker image:

docker-compose up

If you want to proxy calls to the model so that you can add CORs headers, use fastapi_proxy.py: pip install -r requirements_api.txt python fastapi_proxy.py

If you want to use ngrok to expose that proxy to the internet, use: USE_NGROK=True uvicorn fastapi_proxy:app

stabilityai/sdxl-turbo

This is an implementation of the stabilityai/sdxl-turbo as a Cog model. Cog packages machine learning models as standard containers.

First, download the pre-trained weights:

cog run script/download-weights

Then, you can run predictions:

cog predict -i prompt="21 years old girl,short cut,beauty,dusk,Ghibli style illustration"

Example:

"21 years old girl,short cut,beauty,dusk,Ghibli style illustration"

Note:

The model is intended for research purposes only. Possible research areas and tasks include

• Research on generative models.
• Research on real-time applications of generative models.
• Research on the impact of real-time generative models.
• Safe deployment of models which have the potential to generate harmful content.
• Probing and understanding the limitations and biases of generative models.
• Generation of artworks and use in design and other artistic processes.
• Applications in educational or creative tools.