Skip to content // Hype- Text Transfer Protocol

Stable Diffusion is a text-to-image model that will empower billions of people to create stunning art within seconds. It is a breakthrough in speed and quality meaning that it can run on consumer GPUs. You can see some of the amazing output that has been created by this model without pre or post-processing on this page.

Earlier this week the company, founded and funded by Emad Mostaque, announced the public release of the AI art model Stable Diffusion. You may think this is just another day in the AI art world, but it’s much more than that. Two reasons.

First, unlike DALL·E 2 and Midjourney — comparable quality-wise — , Stable Diffusion is available as open-source. This means anyone can take its backbone and build, for free, apps targeted for specific text-to-image creativity tasks.

Alberto Romero – Stable Diffusion Is the Most Important AI Art Model Ever

Don’t believe the hAIp

This morning I read about Stable Diffusion, so this afternoon I thought I’d have a little go. Credit to Stability AI (or more accurately, to their DreamStudio beta) you can go to their site and get immediate free access to a high-quality, fast AI image generator.

The AI is clearly able to create interesting images – just not very specific images. I guess this is the nature of the beast, but I’d hoped for something more. It would be nice if you could enter each part of your prompt and get a sample image back. So you’d type in “In the style of Hergé’s Tintin” and the AI would show you something and say “Do you mean like this?”. This is sort of what’s going on anyway, I guess, but it would be nice to get a clearer handle on what direction the different prompts send the AI.

Without further ado, here are 24 of about 200 images I generated with prompts along the lines of “crocodile with wheels for legs” / “crocodile drives down street cartoon style” / “logo crocodile car funny cartoon” / “driverless crocodile care in the style of Marvel comics” / “crocodile car terrorises Jakarta street” / “crocodile with wheels in the style of Quentin Blake” / “Richard Scarry style crocodile car” / “crocodile rides down Jakarta street in Bajaj warung ojek cat”. (I may have some way to go before I can become a full-time AI wrangler).


See also:
The AI menace that no-one talks about / In which we meet an AI / AI and us: Kevin Kelly on ‘centaurs’ / Open AI’s DALL-E 2 / OpenAI Codex; or, why you might not want to go all in on becoming a full-stack developer

In the unlikely event that you’d like to use one of these images, all image rights belong to me. I’ll say you can use them but please ask first.

I'd love to hear your thoughts and recommended resources...