Some of the links in this article contain affiliate codes. This means that I will earn money at no cost to you if you choose to purchase something. The products mentioned are items I actually own and purchased with my own money.

This past weekend, I downloaded and installed the stable diffusion dataset and code on my M1 Mac using this guide from Replicate. I’ve been playing around with the txt2img script to generate images from natural language prompts.

Here are a few of my favorites.

Monet kitten in a row boat
A kitten in a row boat on the pond, in the style of Monet
Seurat puppy in the woods
Puppy in the woods, in the style of Seurat
Van Gogh pikachu
Pikachu in the style of Van Gogh
Colored pencil pokemon
Colored pencil Pokemon drawing
Skunk pokemon
Skunk Pokemon
Van Gogh cyberpunk hacker
Cyberpunk hacker in style of Van Gogh
Water color pinkie pie
Pinkie Pie in water color
Renoir Mickey Mouse
Mickey Mouse in the style of Renoir
Sonic the Hedgehog computer programmer
Sonic the Hedgehog as a computer programmer
Bill Gates and Steve Jobs oil painting
Bill Gates and Steve Jobs portrait on canvas with oil paint

Keep in mind that these images were generated by the AI using a dataset consisting of around 2.3 billion images from around the web that were used to train the model. The images it creates are unique images that didn’t exist before I typed the prompt.

This is truly amazing when you consider it. Imagine the applications of this technology as it continues to improve. Rather than searching for the perfect image for your website or blog in a stock image library, you could tell the AI exactly what you want and it could generate a new image for you based on your brand guidelines and website’s color scheme.

As with most new technologies, there’s also plenty of room for abuse. Still, I’m excited to see the new opportunities that emerge from AI as it continues to develop at such a rapid pace.

Amazon and the Amazon logo are trademarks of Amazon.com, Inc, or its affiliates.

Tech