If you want to run Stable Diffusion on your own hardware, click here.
Stable Diffusion is a new, free to use, model hosted by Hugging Face, which was produced by Stability AI. This model takes in a text prompt, and produces an image. It was trained on 5.85 billion images. What’s most exciting about it, is it is freely available, and you’re able to run it on your own computer at home!
I’m using a 2080 TI (this is an affiliate link, encase anyone purchases one of these now ancient GPUs ;). This allows me to create an image in under five seconds!
Prompt Writing Tips
The more prompts I write, the more I learn that “more is more”. I find myself briefly describing the desired scene, but using a lot text to describe the quality of the scene itself. For example, instead of writing:
a photograph of an astronaut riding a horse
and getting this:
you can instead write:
a photograph of an astronaut riding a horse, photo realistic, 4k, detailed, vivid colors, excellent lighting
and get this:
Both of these prompts were created using the same seed, 8318, width and height, 512, 50 inference steps, and a guidance scale of 7.5 Yet the differences are substantial.
Example Prompts and Images
Cute and adorable boar wizard, wearing coat and suit, steampunk, lantern, anthromorphic, Jean paptiste monge, oil painting
Notice how different the results are when we even change just one word
Cute and adorable fox wizard, wearing coat and suit, steampunk, lantern, anthromorphic, Jean paptiste monge, oil painting
You can generate beautiful landscapes, these are my favorite photos I’ve generated thus far…
A beautiful castle beside a waterfall in the woods, matte painting, trending on artstation HQ
underwater city of atlantis, mermaids, highly detailed, d & d, fantasy, highly detailed, digital painting, trending on artstation, concept art, sharp focus, illustration, global illumination, ray tracing, realistic shaded, art by artgerm and greg rutkowski and fuji choko and viktoria gavrilenko and hoang lap
512 by 768
50 steps
15 guidance
cloud city, Crepuscular rays, god rays, highly detailed, d & d, fantasy, highly detailed, digital painting, trending on artstation, concept art, sharp focus, illustration, global illumination, ray tracing, realistic shaded, art by artgerm and greg rutkowski and fuji choko and viktoria gavrilenko and hoang lap
512 by 768
50 steps
15 guidance
a hidden oasis, highly detailed, d & d, fantasy, highly detailed, digital painting, trending on artstation, concept art, sharp focus, illustration, global illumination, ray tracing, realistic shaded, art by artgerm and greg rutkowski and fuji choko and viktoria gavrilenko and hoang lap
512 by 768
50 steps
15 guidance
Images of our universe
or even goofy things, like Groot made of our bratwurst
The superhero groot made out of bratwurst
or
giraffe brushing its teeth with a toothbrush, highly detailed, d & d, fantasy, highly detailed, digital painting, trending on artstation, concept art, sharp focus, illustration, global illumination, ray tracing, realistic shaded, art by artgerm and greg rutkowski and fuji choko and viktoria gavrilenko and hoang lap
512 by 768
50 steps
15 guidance
dog knight, portrait, finely detailed armor, intricate design, silver, silk, cinematic lighting, 4k
512 by 512
20 sampling steps
7.5 CFG scale
I hope you enjoyed these images as much as I enjoyed making them!