/sdg/ - Stable Diffusion General

Previous thread: Broken thread chain by retards: |

>Local install
GPU: rentry.org/voldy
CPU: rentry.org/cputard

>AMD GPU Guides
Docker: rentry.org/sdamd
Onnx: rentry.org/ayymd-stable-diffustion-v1_4-guide
Docker on Arch: rentry.org/amddockerarch

>M1 Supported Fork
github.com/lstein/stable-diffusion

>Cloud Hosted Install
Colab: colab.research.google.com/drive/1Iy-xW9t1-OQWhb0hNxueGij8phCyluOh
Paperspace: rentry.org/865dy

>SaaS Sites
txt2img: huggingface.co/spaces/stabilityai/stable-diffusion | beta.dreamstudio.ai | dezgo.com | (NSFW) pornpen.ai
img2img: huggingface.co/spaces/huggingface/diffuse-the-rest
inpainting: huggingface.co/spaces/fffiloni/stable-diffusion-inpainting | inpainter.vercel.app/paint

>Krita Plug-ins
github.com/sddebz/stable-diffusion-krita-plugin
flyingdog.de/sd/en

>Other Guides
Inpainting/Outpainting: rentry.org/drfar
Upscaling images: rentry.org/sdupscale
Textual inversion: rentry.org/textard
Training: rentry.org/informal-training-guide

>In-Depth Usage Info
Any Forums wiki: wiki.installgentoo.com/wiki/Stable_Diffusion
SD Akashic Records: github.com/Maks-s/sd-akashic

>Tools & Resources
Big index: pharmapsychotic.com/tools.html
Artist styles: rentry.org/artists_sd-v1-4 | urania.ai/top-sd-artists | sgreens.notion.site/sgreens/4ca6f4e229e24da6845b6d49e6b08ae7
Prompt sharing: lexica.art/ | krea.ai/
Dataset: rom1504.github.io/clip-retrieval/ | laion-aesthetic.datasette.io/laion-aesthetic-6pls/images

>Finetuned Models
rentry.org/yrpvv

>Related
/h/ /sdhb/: /vg/ /aids/: /mlp/ /ppp/: Official: discord.gg/stablediffusion
NSFW: discord.gg/2ZhBzDGf8u
Training: discord.gg/xVsyrmhQWS

Stability.ai updates: twitter.com/EMostaque

Attached: 07758-210463684-close_shot,_beautiful_dark_catgirl_(long_hair),angry_,wearing_a_heavy_paladin_armor_asymetrical,_(crazy_eyes)_(eyeliner),_small_.png (512x704, 431.07K)

Other urls found in this thread:

github.com/AUTOMATIC1111/stable-diffusion-webui/wiki/Troubleshooting
github.com/AUTOMATIC1111/stable-diffusion-webui/wiki/Run-with-Custom-Parameters
youtu.be/dD9vpO2zaJw
twitter.com/SFWRedditVideos

Oi my niggers, is it better to use this on linux or windows with an nvidia card?

mac

nigger

>How to fix hands

Attached: 00066-237645782-(((beautiful woman))), ((cute)), wearing a traditional european dress, (((big breasts))), model, instagram, very sharp, high det.png (512x512, 383.51K)

Newfag here. I got everything to download and run correctly. However, whenever I try to generate an image, all I get is a black square.

What am I doing wrong?

linux will let you access a small amount more of your VRAM because windows reserves some and won't let CUDA access 100% of it even when it's not using it
But unless you only have 4GB this won't matter, with 8 or more you can do all the resolutions that the model actually produces coherent results for even in windows

(((intricate human hands fingers)))

negative: ((((mutated hands and fingers))))

hope this improves your Booba

Attached: grid-0000-88272135.png (1024x1536, 2.42M)

got somethin cookin

Attached: 1.png (512x512, 368.57K)

>batch size
>79k

h-haha


--no-half --precision full

github.com/AUTOMATIC1111/stable-diffusion-webui/wiki/Troubleshooting
github.com/AUTOMATIC1111/stable-diffusion-webui/wiki/Run-with-Custom-Parameters

Attached: firefox_u5X08IOLNY.png (1031x353, 31.3K)

input:

Attached: king louen leonceur w background.png (1093x958, 488.64K)

what a wonderful technology

Attached: 1638061652188.png (512x768, 483.45K)

Attached: 00065.png (1408x1152, 1.79M)

thank you fren

Attached: 1663276483183322.jpg (493x463, 122.29K)

output:

Attached: 00087-1044865159-award winning digital art by Alexandre Cabanel, 15th century french king, with a crown, majestic beard, blond, long and fair hai.png (1024x1024, 1.31M)

nice

Attached: 1656971395739.png (512x768, 472.76K)

thanks user

Attached: unknown (3).png (1024x728, 1.84M)

how is this original pic called ?

goat.se

Very impressive.

youtu.be/dD9vpO2zaJw

lmao, this shits gonna be making me laugh for a while
man shows user his gaping asshole

Attached: 1644363349978.png (512x768, 516.02K)

paint.net plugin when?

its kinda cool but im having a tough time getting anything but faces
any tips?

Attached: 2.png (512x512, 347.57K)

Why are their thwo threads

more body descriptors

Attached: 1640807995437.png (512x768, 534.92K)

Attached: 00000-~1.png (512x768, 641.77K)

Thanks user. I also found the answer on the Vordy guide. Should have checked that out first.

patrick nagel?

cute
not mine but similar themes

Attached: 1642196626347.png (512x768, 581.75K)

there's no one person who makes all the threads, so people wind up with duplicates.

Attached: stately.png (896x1280, 1003.48K)

>photoshop
linuxbros...

>comments are filled with bots
lol

The krita plugin existed before the photoshop one

why the fuck are my eyeballs spilling. I had them good earlier.

Attached: grid-0003-1603381797.png (1024x1536, 2.16M)

What does the "img2img alternative test" script on Voldy's do?

>try the waifu model
>tits are way more massive

Attached: 00656-123458-(((large breast))), (dynamic posing), blond girl, serious, highly detailed, 4k, (painting), ((Albert Lynch)), smile, young women.png (512x576, 370K)

sometimes I like when they get weird

Attached: 1632208964317.png (512x768, 612.32K)

They broke the thread chain by not including the previous thread in the OP

Next thread should start with:

Old breads: >> 88632331

Attached: out-76.png (512x512, 505.9K)

Attached: 1663267578749.png (1536x768, 2.66M)

>paint.net
PHAHAHAHAHA

Attached: 1660292472516.gif (292x292, 1.44M)

it's better than gimp

It's tuned on an anime booru, so yeah.

Attached: output 1.png (512x512, 605.14K)

I'm gonna run out of storage, I can't bring my self to delete anything

now sd upscale it, then select a tiny spot and infill a wally

Notice how it completely ignores the golden collar?
Is there a way to force it to incorporate the things it's throwing away?

Attached: 00162-242291913-sexy beautiful woman, goth makeup, realistic, sharp, 4k, 8k, (((holding big breasts))), (((intricate human hands fingers))), hou.png (512x512, 331.94K)

I would agree.

nice

Attached: 1649171652479.png (512x768, 588.4K)

And that's a good thing

she seems to be missing a nipple

i like it this way, how can femoids compete?

Attached: out-1.png (512x512, 619.78K)

really good

No idea how to enforce it, I was expecting the AI to draw the necklace like pic rel.

Attached: louen transparent.png (940x940, 1.04M)

>how can femoids compete?
they cannot.

Attached: 1646880894287.png (512x768, 595.11K)

How well can it create fictional characters from text, Thecla from botns for instance?
I'm curious what it will spit out for a beautiful exultant

Attached: out-2.png (512x512, 578.97K)

Jesus that's great

it draws my oc pretty well just from text descriptions

That's not AI btw, it's a portrait from my warhammer mod

Attached: yuri generator.png (704x448, 445.12K)

is switching to the waifu model just a matter of swapping out the ckpt file or is there other faggorty involved? I thought I got them titties out of my system yesterday, but my brain just got short circuited again by

First describe the oc
Then post the image you got

Maybe that's the reason why it's so soulful

excellent, how can I learn this black art

Attached: 3994139504_terrifying_animal_at_the_end_of_a__3d_tunnel__horror__hq__startling__3d_art__colorful__cracked.png (512x512, 534.81K)

SOMEBODY CONVERT A CHRIS CHAN DRAWING ASAP

no, make your own OC
or use waifulabs and then describe the output in plentiful detail to waifu diffusion

wrong pic, just fix it and they are ready to go, the problem is the resolution

Attached: download (37).jpg (1527x1022, 1.49M)

I have no computer

Thanks, it's just a kitbash from existing assets and a bit of upscaling on his textures using ESGRAN

Attached: Screenshot (1125).jpg (1920x1080, 891.32K)

One day someone will post the real one and anons will click on thumbnail thinking it's AI generated image.