In this tutorial you'll learn the easiest way to deploy Stable Diffusion to production on serverless GPUs.
This deployment demo is completed in about 12 minutes (minus build time), making this one of the more efficient methods out there to deploy Stable Diffusion. We walk through all of the steps required to deploy, from creating your GitHub repository to actually running an inference in production with your model on serverless GPUs. Enjoy!
If you are looking for ideas of what project you could build, check out our list of badass Stable Diffusion projects already built and new ideas you can steal!
What is Stable Diffusion?
Stable Diffusion is a cutting-edge text-to-image machine learning model developed by stability.ai that generates images from text.
One of the key innovations with the Stable Diffusion model is its speed. The model can run on consumer-grade GPUs while producing high quality visual art. Stable Diffusion requires 10 GB of VRAM on consumer GPUs and will generate images at 512x512 pixels in a few seconds.
The goal of being able to run on consumer GPUs is to democratize image generation by enabling researchers and the public access to this high performance model.
Tutorial Notes & Resources:
We mentioned a few resources and links in the tutorial, here they are.
- Creating a Banana account (click Sign Up).
- Access to our Stable Diffusion serverless template repo.
- View the Stable Diffusion model on HuggingFace.
- How to create an access token on HuggingFace.
In the tutorial we used a virtual environment on our machine to run our demo model. If you are wanting to create your own virtual environment use these commands (Mac):
- create virtual env:
python3 -m venv venv
- start virtual env:
- packages to install:
pip install banana_dev pip install diffusers pip install transformers
What did you think of this tutorial? We'd love to chat if you have any questions or want to talk shop about Stable Diffusion. The best place to reach our team is on our Discord or by tweeting at us on Twitter. Do you have a machine learning model you'd like to see a deployment tutorial for? Hit us up!