During the last few months, my interest in AI image generation has grown a great deal. It started with me reading about DALL-E
, and realizing how much potential this has for creativity, and was renewed with the release of stable-diffusion
under an Open Source license.
Compared to DALL-E, stable-diffusion has very important advantages:
- It’s free
- It’s open source, so it can be inspected and modified
- It can be downloaded and used on any computer with a recent GPU
Since the release of stable-diffusion, there has been an insane amount of innovation from the open source community. User interfaces have been released, tutorials, modified versions, guides on how to disable filters, etc.
It’s really an understatement, but the pace of innovation there is astounding. And it will only keep accelerating, as more and more people discover it. People are already experimenting with video.