Having spent several months reviewing the capabilities of the new crop of online AI-based image generation tools, I decided to take a look at the core software. There were a couple of reasons for my decision.
- It opens up the inner workings of these amazing creative tools.
- Running on your own PC is free unlike most of the current sites.
- There is greater control and scope over the creative process.
So, I chose to install Stable Diffusion on my Windows workstation. There is a requirement for a reasonable NVidia GPU so this PC, being configured for a bit of gaming, was ideal.
Stable Diffusion is an AI model that can generate images from text prompts or modify existing images with a text prompt, much like MidJourney or DALL-E 2. It was first released in August 2022 by Stability.ai. It understands thousands of different words and can be used to create almost any image your imagination can conjure up in almost any style. There are two critical differences that set Stable Diffusion apart from most of the other popular AI art generators, though; as noted above, it can be run locally on your PC and it is an open-source project.
There are several ways to install Stable Diffusion on a Windows PC. One way is to use the official standalone installer from Stability AI. Another way is to use the NMKD installer, a third-party package for Stable Diffusion that takes care of all the processes and has an easy-to-use UI. Lastly, if you prefer, you can choose to install Stable Diffusion manually, where you will have to download and install dependencies on your own and then clone the web-ui so you can use it on your PC.
I chose the latter as I’m comfortable with installing software and it gave an even more in-depth view into how the tools fit together.
Here’s how to install Stable Diffusion manually:
- Download and install Python on your PC.
- Download and install Git on your PC.
- Clone the web-ui from GitHub.
- Download a Stable Diffusion model file.
- Set up web-ui.
- Run Stable Diffusion.
For more detailed instructions on how to install Stable Diffusion manually, please visit this link.
The install went remarkably smoothly. So watch out for future posts on how my experiments go.