/sdg/ - Stable Diffusion General

Old Bread: >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

>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

>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: hgsd.png (512x512, 672.2K)

Is this the chucke cheese thread

Outpainting OP

Attached: Famous paintings mosiac.webm (540x360, 2.25M)

Attached: ww.png (512x704, 626.25K)

Dinosaurs

Attached: Jurassicpark.webm (640x448, 2.88M)

i'm getting close

Attached: file.png (512x512, 464.27K)

Some of the shit I'm getting looks better than what you see in fashion shows. Amazing
yes I watch fashion shows they're enjoyable

Attached: 1.png (448x640, 382.92K)

I assume this is voldy's white whale right now.

okay ive been messing around for a few days with generating esoteric pdfs. any prompts been posted to make stuff like this way better? i noticed one thread had some stuff posted but no prompts. share prompts please

anyone want to make our own pdf to scare schizos with?
>reply to this post with a page and prompt
>post number determines page number

Attached: demon anatomy journal page.png (512x768, 881.66K)

artists are fucking done

It's open source so it's not going to be hard to implement

Oh no no no

Attached: 1663261070454.jpg (1614x814, 121.06K)

CFG Scale
What is this? How do you use it?

Attached: 00020-3941429492-.png (512x512, 351.68K)

whoao

prompt?
didn't work
worked, but the prompt was [naked:clothed:x]

Attached: xy_grid-0027-30891209002.png (1534x580, 1.33M)

>Stability.ai has a home page mentions Harmonai - an organization that makes AI driven tools for music production
Any info on when they will be releasing anything? If this is what current AI can do with images, then I can't wait to hear what music made with similar technology will sound like.

bros...she beautiful.....

I might be sympathetic to artists who work hard but Getty needs to die
fucking leeches

if i follow the VOLDY RETARD GUIDE (GPU) step by step, will i be later able to use the AUTOMATIC1111 version of Stable Diffussion?
i don't understand how these things work yet but i want to generate all kinds of stuff.

holy fuck

it's a generic artstation cumdumpster

Attached: 843029123.png (1024x1536, 1.84M)

Artists on sucide watch!
Emo banned nationwide!
Sale of oil pastels limited to one set per customer!
Michael's posting record losses!
Bob Ross Foundation closes doors permanently!

Finally, those filthy degenerates can go wallow with the rest of the poors.

Attached: swamp trump.png (512x512, 323.61K)

Attached: xy_grid-0233-4214474433.jpg (2944x1615, 645.31K)

this guide instructs you how to install AUTOMATIC111
just fucking read it instead of asking dumb questions

I love outpainting.

Attached: Paramount Pictures Outpainting.png (2240x2432, 3.79M)

those are the same versions retardkun

thank you sir.
i hate reading. sorry

Attached: 1.png (448x640, 380.06K)

Attached: elsa (14).png (512x640, 502.33K)

show me how to make peepo images

Attached: 173967251463.jpg (810x802, 96.86K)

Here are some comparison images between different AI gens.

Attached: AI_Comp1.png (3072x1024, 1.37M)

What we really need is an AI trained on filling in transparent pixels rather than using the stretch pixel technique (which is causing the repetition artifacts).

How closely it follows your prompt

1 = The AI follows your prompt loosely (you can kind of do what you want AI)

16 = The AI sticks to your prompt like glue (it's my way or the highway!)

...

Porn industry is over.

If i have a 1280x768 image, how do i go about upscaling it with SD upscale? I tought it automatically doubled the res but it's stayed the same.

this is really cool.

Attached: grid-0012-964974023.png (1024x768, 1.41M)

that's a man

Which web UI repo is better right now, hlky's or AUTOMATIC1111's?

Attached: 424-4249790_transparent-emoji-thinking-png-dog-thinking-emoji-meme.png (920x977, 50.86K)

>legible and sensible words in midjourney
is that real

Tick Denoise up from .5 in .01 increments until pepe becomes a stack of bullshit.
It does help if your prompt includes things that are the same color as the original image.

Attached: pipe.png (768x768, 593.62K)

Attached: AI_Comp2.png (3072x1024, 1.77M)

>cocka

Attached: 1652504100341.png (600x692, 381.47K)

midjourney seems to be the most original. stable diffusion is the best looking. then dall-e 2 and then the rest. craiyon looks the funniest tho. flow just merged the stuff.

Dude it’s over. Every single regret I ever had about never drawing or painting (which were negligible to begin with) has completely evaporated.

4 passes with
>3d environment, realistic, sci-fi, building

Attached: compare.png (1023x512, 2M)

Attached: AI_Comp3.png (3072x1024, 1.29M)

Now make a photobash with Godzillas, cyber Stalin, UFO dogfighting in the sky, and some other insane shit happening outside of the classic Paramount frame

thank you for the horrifying prompts, you're doing god's work

oh my god that is real

Some of these comparisons aren't fair or intelligently designed. Glass burger can be interpreted in a thousand different ways. Glass in a burger? Burger made of glass? Glass next to a burger?

keep em coming user, these are the most beautiful women to not exist

Texturing your occultist lab in an indie horror game has never been easier.

Attached: (((Arthur Hughes))). Alasdair McLellan 129.png (448x640, 394.58K)

Attached: AI_Comp4.png (3072x1024, 1.87M)

Anyone up for some Stargate adventures?

Attached: daniel.png (512x512, 350.85K)

concept artists gonna freak

Attached: 1648991612664.png (256x256, 53.93K)

cocka kooa

it's so quiet with all my miners turned off

Yeah but selling it will never be harder because there's going to be a million people doing the same thing as you

This is a great idea that someone should hop on.

Attached: AI_Comp5.png (3072x1024, 2.1M)

How old is she?

hlky is behind on quite a few features
most everyone uses automatic now

80/20 rule times infinity. People who make shit won't stop making shit because of the AI.

Attached: elsa (15).png (512x640, 567.67K)

What irks me the most is that you probably recouped the costs long time ago. I hope you die in your sleep tonight.

If this board is any indication, I'll only have to compete with the top 5% same as usual.
Most of you fucking suck at this.

Attached: dieselpunk_tower.png (512x512, 488.46K)

Love the Monkey scientists around the fire in Stable Diffusion. Period correct.