Like the rest of the world, I've been playing around with AI generated art. In my case, I've been using Stable Diffusion UI. Stable Diffusion UI is an easy to install variation of Stable Diffusion that also includes a GUI. I was able to get running on both Windows and Linux with minimal effort. While the included GUI is a nice convenience, the real appeal to me of this version was the fact that it was very easy to install.
To get speedy results you need a decent GPU. However, you don't need a GPU to use Stable Diffusion UI. It's just that if you don't have one and use only the CPU, each image will take 30 minutes or more to produce (depending on how fast your CPU is, the settings you use, and various other factors). However, this is open source software so you can generate as many images as you want for free. So even with limited hardware, you can still produce unlimited images as long as you are patient.
I believe I started out with Steampunk themed imagery and I am back to that with this post. The steampunk aspect is somewhat understated in this portrait but I still like how it turned out. The fact that the visible hand came out reasonably normal is a nice bonus. Very often with Stable Diffusion and AI generated art in general hands do not turn out well.
One interesting observation... In general, I have been trying to generate portrait style imagery. Recently, the images being produced by my prompts included two or more individuals or the same individual duplicated in some fashion. It turned out that this correlated with increasing the resolution of the generated images. I knocked the resolution back down and I'm not having that problem any longer.
