A Stable Diffusion Guide to install and run everything on your own PC

[Hongfire Survivor]

So Novel AI's model was leaked to the interwebs and this handy guide also decided to pop up:

You must be registered to see the links

An amazing write up by Automatic1111.

Please read everything as it covers, well, everything. You'll have a few different models to choose from including Novel Ai's.

You will need to experiment with Inputs, Sampling, and the CFG Scale, among other parameters. If you are NOT specific in your prompts, expect nightmare fuel. You have been warned.

I'm not an expert on this myself, so research (Google) and get better over time.

Can also confirm it works as intended. Mods, feel free to move the thread or nuke me for breaking a rule, I didn't see anyone post this here from a quick search.

Also a now deleted Blog about experimenting with different Artists:

You must be registered to see the links

Please note: Use ((())) instead of {{{}}} to tell the prompt what to focus on.

 

[Hongfire Survivor]

Also a guide on making expressions and other tweaks:

You must be registered to see the links

 

[hentaipt]

help?

 

[Hongfire Survivor]

help?

From the guide, not me:


--RUNNING ON 4GB (And under!)--

These parameters are also useful for regular users who want to make larger images or batch sizes!
It is possible to drastically reduce VRAM usage with some modifications:

• Step 1: Edit webui-user.bat
• Step 2: After COMMANDLINE_ARGS= , enter your desired parameters:
  Example: COMMANDLINE_ARGS=--medvram
• If you have 4GB VRAM and want to make 512x512 (or maybe larger) images,
  use --medvram.
• If you have 2GB VRAM,
  use --lowvram

If you are getting 'Out of memory' errors on either of these, add --always-batch-cond-uncond to the other arguments


NOTES:

• If you get a green/black screen instead of generated pictures, you have a card that doesn't support half precision floating point numbers (known problem on 16xx cards):
  You must use --precision full --no-half in addition to other flags, and the model will take much more space in VRAM
  If you are using a .vae file, you must also add --no-half-vae
• Make sure to disable hardware acceleration in your browser and close anything which might be occupying VRAM if you are getting out-of-memory errors, and possibly remove GFPGAN (if you have it)

 

[arsioxy]

Hopefully ppl from the SD-web-ui's repository add support for AMD on Winlel or something. It works fine on Linux but it was such a pain in the ass to set up, like holy shit I had to jump through a bunch of hoops because AMD ships a broken deb package.

 

[TGIL]

I wish someone would make all in one software for the less talented. However I did find some and they didn't work for me either. There's this w/ an updated list.

You must be registered to see the links

 

[Hongfire Survivor]

I wish someone would make all in one software for the less talented. However I did find some and they didn't work for me either. There's this w/ an updated list.

You must be registered to see the links

That is a lot of options.

I heard good things about this one:

*PICK\* (Updated Sep. 5, 2022) GitHub repo

You must be registered to see the links

by cmdr2. Windows program and Linux program. "Easiest 1-click way to install and use Stable Diffusion on your own computer. Provides a browser UI for generating images from text prompts and images

 

[ThreeNine]

Are there any guides for making videos out yet?