Stable Diffusion Gets Public Release
Stable Diffusion has been released to the public, including a Hugging Face demo internet users can go play with right now.
Created on August 23|Last edited on August 25
Comment
Earlier this month, Stable Diffusion had a beta release meant for researchers, but now, as promised, it's getting it's full public access release. The public release gives everyone access to model weights and all the code needed to run it, as well as an easy to use demo app built with Gradio hosted on Hugging Face Spaces.
Like when Craiyon (at the time, DALL·E mini) was unleashed to the masses with Gradio's new demo creation tools, it wouldn't be a surprise if the internet denizens took it upon themselves to swamp the servers hosting Stable Diffusion's demo.
How to use Stable Diffusion?
The easiest way to get right into playing with Stable Diffusion is at Dreamstudio's beta location. Using the demo will let you create batches of up to four images based on your input text, and for me usually takes around 5 or so minutes to complete at time of writing.
For those who want to run Stable Diffusion on their own terms, model weights and setup code are available on the Hugging Face model card page as well as the GitHub repository. The public release of this model does come with a license agreement, but other than what's outlined within it, you're free to do what you want with it.
Find out more
Add a comment
Tags: ML News
Iterate on AI agents and models faster. Try Weights & Biases today.