Working with sizes

You can specify the resolution that you want, but caveat — AI models react unexpectedly to free-size resolutions.  To account for this, we created shortcuts that render the most compatible dimensions without burdening you to remember the exact pixel numbers.  They’re easy to use:

For beginners especially, we highly recommend starting with these.

Venturing into specific resolutions

You can use /size to specify a manual number, like this:

/render /size:830x1024

For best results, start with a resolution that can later be AI-upscaled 4x, instead of trying to do HD right away.

The recommended resolutions are: 

768, 768

Why not go straight into a larger size?

Glitches. Multiples of 64 have the highest compatibility, but may result in unpredictable issues like multiple heads, cloned characters, etc. because the training data is based on small squares of 512×512, or 768×768. It would also take an insane amount of VRAM, it’s just not practical.  Target one of these resolutions and then use AI upscaling to get the best-looking pictures.

Learn about the three modes of Facelift (and how to disable face retouching)

What are Steps in Stable Diffusion?

/steps refers to the number of render passes that the server takes to arrive at the final image. More steps can mean a more detailed image, while fewer steps can be considered a rough sketch, though in some cases less it can result in an “overcooked” result, especially when drawing line art like Anime.

Usage:   /render /steps:15  and your prompt etc

15 steps is considered on the low end, and 200 steps is considered whoa.  As seen on popular prompts on civitai, you only usually need between 15-35 steps for a beautiful image.  There’s no need to do crazy step counts, it’s slower.

Steps + Images in one handy shortcut

Our community shares a large multi-server instance of Stable Diffusion. We want everyone to wait around the same time, and as little time, as possible.

To balance this, we suggest using these shortcuts.  Each one takes roughly about the same time to render, but you can get more images back.

/render /steps:wayless — returns 9 images

/render /steps:less — returns 6 images

/render  — returns 5 images (default 50 steps)

/render /steps:more — returns 3 images

/render /steps:waymore — returns 2 images  (experimental, upper limit)

Suggested workflow

We talk to many great AI creators — and they throw away 99% of their stuff. This is the way they work:

Begin with /render /steps:wayless to get a lot of ideas fast, then when you start to arrive at poses and images you like, use /showprompt

Showprompt is a command that gives you all of the information you need to regenerate an image.  Armed with the one you like, modify that prompt with more steps and you’re arriving at your destination faster instead of rendering slow responses every time that you may not want to keep anyway.

Showprompt is the fastest way to find out an image’s Seed.