SDXL Turbo is an adversarial time-distilled Stable Diffusion XL (SDXL) model capable of running inference in as little as 1 step.
This container runs the SDXL Turbo model as a web service using FastAPI. The inference code uses HuggingFace Pipelines. The container is available on the AWS Marketplace from Stability AI.
Once you have pulled the Docker image using the instructions on the AWS Marketplace listing for the SDXL Turbo container product, run the container to start the server:
Run on a cuda device with:
$ docker run --gpus all -p 8080:8080 -t <image-name>
To test the model:
cd sdxl_turbo
pip install -r client_requirements.txt
python client_request.py