Stable Diffusion Rentries + Simple Installation Tutorial!

Here's how to download Stable Diffusion WebGUI

Requirements:
At least 4GB of dedicated memory
At least 10.5GB of hard drive space for everything
+more if you download custom Dreambooth models

Unlike InvokeAI, which requires a great NVIDIA gpu, WebGUI can run on a wide variety of NVIDIA gpus.
I'm guessing you need a discrete GPU separate from your processor.
I have an old GTX 1050 graphics card, and it still works.
If yours is any less powerful, good luck.

Download:
Install Git for Windows,
https://git-scm.com/download/win
Python version 3.10.6 specifically,
https://www.python.org/downloads/release/python-3106/
the actual WebGUI fork,
https://github.com/AUTOMATIC1111/stable-diffusion-webui
and the model which has all the images trained on it
https://huggingface.co/runwayml/stable-diffusion-v1-5

Make sure Python is selected for "Add Python 3.10 to PATH"
Make sure to always download the 4.2GB "v1-5-pruned-emaonly.ckpt", you don't need the higher storage model

Steps:

Create a folder, name it Stable Diffusion or any other name
Open the Windows Start Menu, and search up the Command Prompt
Open Command Prompt with "Run as Administrator"
In the Command Prompt, copy and paste the path to your folder
cd C:\Users(NAME)\Desktop\Stable Diffusion

Enter the command

⎗

1	git clone https://github.com/AUTOMATIC1111/stable-diffusion-webui.git

Move your ckpt file into the models folder
ex: Stable Diffusion > stable-diffusion-webui > models > Stable-diffusion
It has "Put Stable Diffusion checkpoints here.txt"
In the stable-diffusion-webui folder, open the "webui-user.bat" file with a notepad or notepad++

Copy the file path to the installed Python's exe into "set PYTHON" and make sure webui-user.bat looks like this:

⎗
✓
git pull
@echo off

set PYTHON= (COPY YOUR PATH HERE)
set GIT=
set VENV_DIR=
set COMMANDLINE_ARGS=

call webui.bat

Double click "webui-user.bat" and wait until you can copy a local URL

Running on local URL: http://127.0.0.1:7860

Open a web browser and enter the 127.0.0.1:7860 URL.
Later on, right click the "webui-user.bat" and make a shortcut to the Desktop

Send to > Desktop (create shortcut)

Holup, what's this loRA and ControlNET thing?! idk how i can even start figuring these things out, but ill try.

I can barely differentiate between Dreambooth and Textual Inversion

Models:
Look through HuggingFace or civitai

Some suprise tools that will help us later:

This was my old thing

-

Old stuff

here
https://mattvidpro.com/
https://docs.google.com/document/d/173yCd2ZWYMEUQMK-hCD7zCeP1v9gUOVCwB-hequWZYA/edit#heading=h.wn663pqlkwoj
https://reddit.com/r/StableDiffusion/comments/wqaizj/list_of_stable_diffusion_systems/
https://reddit.com/r/StableDiffusion/comments/xvhavo/made_an_easy_quickstart_guide_for_stable_diffusion/

https://beta.dreamstudio.ai/home
https://huggingface.co/spaces/anzorq/finetuned_diffusion
https://huggingface.co/spaces/stabilityai/stable-diffusion
https://playgroundai.com/
https://dreamlike.art/
https://getimg.ai/
https://pollinations.ai/
https://mage.space/
https://deepai.org/machine-learning-model/text2img
https://www.artbreeder.com/beta/browse
https://colab.research.google.com/drive/1_8wPN7dJO746QXsFnB09Uq2VGgSRFuYE#scrollTo=1HaCauSq546O
https://dbzer0.itch.io/stable-horde-client
https://replicate.com/lambdal/text-to-pokemon

prompt
https://lexica.art/
https://libraire.ai/
https://openart.ai/
https://openart.ai/promptbook
https://publicprompts.art/
https://promptomania.com/
https://www.reddit.com/r/StableDiffusion/comments/x7mbxa/a_comparison_of_the_top_500_sd_artist_styles_see/

https://huggingface.co/spaces/Gustavosta/MagicPrompt-Stable-Diffusion
https://www.coolgenerator.com/sentence-generator
https://promptomania.com/stable-diffusion-prompt-builder/

artists
https://rentry.org/artists_sd-v1-4
https://proximacentaurib.notion.site/e28a4f8d97724f14a784a538b8589e7d?v=42948fd8f45c4d47a0edfc4b78937474
https://proximacentaurib.notion.site/2b07d3195d5948c6a7e5836f9d535592?v=affb330abb7d46bdbc2aa6d0917a656f
https://sgreens.notion.site/sgreens/4ca6f4e229e24da6845b6d49e6b08ae7?v=fdf861d1c65d456e98904fe3f3670bd3
https://docs.google.com/document/d/1SaQx1uJ9LBRS7c6OsZIaeanJGkUdsUBjk9X4dC59BaA/edit#

upscale/fix
https://replicate.com/jingyunliang/swinir
https://replicate.com/nightmareai/real-esrgan
https://replicate.com/nightmareai/latent-sr
https://replicate.com/nicholascelestin/real-esrgan-nitroviper
https://replicate.com/tencentarc/gfpgan
https://imglarger.com/
https://clipdrop.co/image-upscaler
https://ai.nero.com/image-upscaler
https://huggingface.co/spaces/akhaliq/SwinIR
https://deepai.org/machine-learning-model/waifu2x
https://github.com/n00mkrad/cupscale
https://colab.research.google.com/drive/19euI_7GAgbvMoZsuPj9SZseDeuFnwBj8#scrollTo=_hrsrSgFpB1V
https://clipdrop.co/relight
https://docs.google.com/document/d/1fv-a3Mr-k13SrgsuvjFt8FsppMsMPOrnPEfHeRfBNyE/edit

my other rentrys
https://rentry.co/StabilityAI
https://rentry.co/DreamStudio
https://rentry.co/sddiffusion

https://rentry.co/DALLE
https://rentry.co/DALLE2
https://rentry.co/DALLE3
https://rentry.co/DALL-E2
https://rentry.co/DALLEmini

https://rentry.co/midjourney (oh my god i fucking lost the code)
https://rentry.co/MidjourneyV4

https://rentry.co/aitestkitchen
https://rentry.co/dreambooth
https://rentry.co/GregRutkowski
https://rentry.co/latentdiffusion
https://rentry.co/mattvidpro
https://rentry.co/neurallove
https://rentry.co/Painthua

I'm sorry no other AI's posted here, but I'd love to post anything as FREE and POWERFUL as stable diffusion

https://discord.gg/stablediffusion
If I dont have what you need go look at other more established creators idk

=

I'm going to revamp this to add some visuals, it looks like the traditional artists are begging for this practice to not appear on social media, to have an opt-in system, and to get the reassurance their jobs and hobbies aren't crumbling like the world around them

It is good that

-People that only want to decorate, conceptualize, and express their ideas in real art
-People point out the lack of coherency or ease of use in Stable Diffusion or maybe Midourney
-People feel a little betrayed with that one "AI art won 1st place in an art competition"
-People want other forms of low effort mediums and memes to not get drowned "low-effort" AI art
-People easily understand Stable Diffusion doesn't match the coherency of portraits and sketches of right now

But it is very unfortunate and undeniable that some people
-Pray that laws and legislatures will strike down on this extraordinary software
-Argue that the camera and digital art make great strides in innovation just because an AI is exponential
-Claim that only this medium of art can spread hate, fear, misinformation, and instant gratification
-Want to eradicate prompt artists in the head like they're the scum of the Earth
-Quiver over the idea that their hobbies or incomes have instantly disappered with the flick of a switch
-Whine with the lack of consent in detailed art added to the datasets

Just because I like AI art doesn't mean I want to replace all traditional art and jobs with Stable Diffusion,
and demonize the collective backbone foundation of this software.

Likewise, just because a few of you hate AI art doesn't mean you should want to replace the communities of inspiration-less & innovation-lacking users with normal art,
demonizing the future tool that will remove the creative blocks and time constraints of expressing ideas.

I may not change your minds, and you don't even have to read what's below here. I'm just writing my compiling and fixing up some stuff I'm thinking of.

v (arguments that need some cleanup and sources) v

Hopefully in the future, projects like this will not only continue to improve, but gain a wider range of acceptance. Unfortunatley, there are some people who lurk around that believe this tech will either be an insignificant fad like crypto-currency, or a dystopian tool that would remove social media's "moral principles" and bring ruin to every legitamite and amazing human artist. From book covers to propaganda, its unfortunate people believe in a world of traditional media, conservative and hesitant, to communicate our ideas, visions, and concepts. Self driving cars and discriminatory job application AIs are the ones that can truly ruin lives, and yet vocal minorities are stuck on this?

"This AI is going to end concept art for my movies and games"
Stable Diffusion cannot tell your story and give your feelings. Only you, the artist, can.
Just because it can make amazing characters, worlds, creatures, weapons, and vehicles doesn't mean it can follow your world-building.
Stable Diffusion is not able to perfectly replicate your own vision of a project.
It can make all the Greg Rutkowski and 3D Pixar renders it wants, but only you can train it to execute your style.
Unless your board game, video game, or movie has little-to-no stylization, your OCs and fan universes are not going anywhere.

"This AI is stealing my art style"
Unless you have a hundred public commisions accurately labeled, you won't all of a sudden have strangers generate your artwork.
Assuming someone really wants to steal your style, it would take a great effort to expand the AI's capabilities.
Off the top of my head, there are small projects popping up like Waifu Diffusion, but as for now, your unique designs are still out of reach

"This AI is plagiarizing my favorite artist's work!"
Unless that artist can provide his or her prompt, the version they are on, their seed, their cfg scale, and what AI they are using exactly, that Twitter user is a liar trying to get attention. All they did was use img2img to generate an AI's take on a variation of the original image. If not, a troll could have done img2img for them, or plagiarized their work by themselves.
Also, why the hell are you blaming the software?! If you catch someone tracing an animation or using the bucket tool on someone's work, do you bombard Photoshop and Procreate?

"This AI doesn't have my permission to use my art"
https://laion-aesthetic.datasette.io/laion-aesthetic-6pls/images
Even if Stable Diffusion truly has your art from LAIONs 5 billion potential images, the random noise of pixels getting denoised multiple times wont be able to utilize the same subjects, details, or effects in the same shape like a pattern. The AI doesn't learn to replicate and copy-paste parts of similar images, it actively learns to figure out what each word means and puts their best representations of them in different frames and scenarios. It doesn't just have one image for each possible artist, one image for each possible celebrity, and one image for each possible painter; it has to blend them all together in a 512 by 512 area.
Do you know who else takes inspiration from the art at their fingertips, to recreate and mash concepts that have been made in the past, into something original and never made before? lol

"This AI requires no skill"
That's okay, from personal experience, prompt-building is a casual exercise that's just about as hard as building a Lego set. And just like building Lego, I understand that there are many experts around the world that can do great things by themselves and many different levels of projects that can be accomplished. Lego and art as a whole are flexible mediums of art, and there is totally no risk in there being easy shortcuts that can get you to spend less time, effort and resources.
Unlike other past-times or discussion topics, online AI like Stable Diffusion is beautiful in a way where you don't need to be good or reach a goal to feel accomplished. Just browse a little of Discord, browse the catalogs of varying art styles, and bam, you just dreamt up a mechanical futuristic mech with red and white accents. Whether we watch other people or partake ourselves in copying and pasting (4k 8k hd highly detailed by Greg Rutkowski) into every prompt, it doesn't take a genius in order to fascinate others or make others laugh.
Things can get more complicated in the future with unforeseen challenges and critics, or get easier with millions of people improving the ease of accessibility. But if you don't want your skill level to stay stagnant and to grow dependent on this concept commission bot, good for you.

"This AI and us current artists cannot work together, its impossible!"
Using more tools in your profession or past time doesn't make you weak or slow, it will help you adapt faster and faster in our changing world, and for anyone who insists on their followers to remain pure, they will simply not catch up. There's only Discord bots and Photoshop Plugins right now, but with all of humanity's art stuffed into 8 to 10 GB, you can ensure that when you're not in the best headspace right now, or are physically unable to use a normal writing utensil, the art just won't stop. When life rains hell on you working artists, AI may not be the umbrella to end it all, but its still a start in persevering through the storm.
Even though I said earlier AI cannot make movable, realistic spaces and perspectives, this doesn't mean a little photoshop magic can't help make the impossible possible. Whether you're making a banquet, a planet, a species of monster, or a photograph to look virtually lifelike, all it takes is hitting the same button over and over again to imagine 4 or 9 usable concepts at a time. Once you find enough meaningful images, you can piece your favorite details of them to make your grand project, and instead of ripping your hair from not knowing what to do, you can reserve that energy and perseverance for finalizing and revising art as fast as you want. AI may make single simple images of a single subject as abundant as oxygen, but when it comes to making distinctive and unique art for a creator collaboration or job, you don't have to worry.
No more memes where "WHEN YOU WANT TO MAKE A {---} BUT YOU CAN'T DRAW, CODE, COMPOSE, EDIT, OR VOICE ACT", because now anyone with a keyboard and mouse can now focus on what hasn't been made easy with free artificial intelligence yet.

"This AI is going to end democracy like Facebook is doing right now, so much misinformation!"
Social media may also use their algorithms to hook their users, and diffusing many images may form addiction/gambling fears, but Stable Diffusion isn't interested in selling your data to advertisers and profiting off your interests. If Stable Diffusion, Midjourney, or DALLE 2 wanted to profit from massive companies needing more "artists", they would be doing it themselves, as it doesn't require roping us users in. But no government so far has expressed any interest in hijacking the next election with blurry images of politicians. If you need a bottomless source of media, you can always use actual news reports, with their 24/7 model milking out as much fear as possible.
Politicians prosper under the guise of religion, using conspiracy theorist content creators, not regulating their questionable representatives Tweets, and displaying blocks of texts.
Political parties A political party doesn't have to hire their own artists to work for them, as their followers can do all the work for them

Cant wait until you can see text to image AI that has occurred months after this release. Many disgruntled artists are worried of this innovation of Stable Diffusion and Dreambooth as "useless", "job-stealers", "exploitative", but most importantly of all, "stealing". Hopefully you or someone else with a high enough platform or expertise in presentation can increase its acceptance.

All these artists here are acting like Nintendo torwards Pokemon fangames; punishing / harassing anyone because refrencing their precious brand or style will make them lose money. Just because you can mass produce artwork doesnt mean you can replace human creativity. You can't force new artists or machines to purposely block out inspirations for "PeRmIsSiOn", and you can't force consumers to only accept high-effort/ethical/human-made art (search up the song "Copying isn't theft" if you haven't heard of it yet)

I just left dropped off following AI news for a short while, and I'm so excited with these developments in AI that can come in the following years or even sooner. Even with the current constraints, I love how anyone would soon have the potential to make any image, video, sound, or bot without the limits of time, energy, expertise, or imagination. I pray that hopefully in the future, projects like this will not only continue to improve, but also gain a wider range of acceptance and funding.
Unfortunatley, there are some people that believe in a world of traditional media, conservative and hesitant, to communicate our ideas, visions, and concepts. Its pretty clear this tech is not being kept under a restrictive monopoly, that it can coexist with current amazing artists, and anything bad that can be done here has already been done countless times in the past (racism, propaganda, defamation, erotic content, low effort works)

It's hilarious some people automatically believe EVERY human work of art is more enjoyable than an AI's, and therfore the works Midjourney and Stable Diffusion shouldn't see the light of day in communities past here, with complaints of "low effort" and "human focus" scattered around.

Its amazing to see how our view of modern art evolves and gets interpreted, but it reminds of the so-called critics wishing to end the momentum from Stable Diffusion and Midjourney.
Imagine the "artists" judging AI art in the future, with its being even newer, more niche, and easier to reproduce without the sweats of labor.
"Ah you see, this dream of a distant galaxy highly surreal matte painting by Greg Rutkowski may captivate attention and thoughtfulness, but this is the work of a dirty fat prompt builder! He's just piggybacking off a machine that has scanned a thousand works of real art of the night sky! Who gives him any right to post his low-effort crap in our sacred sanctuary of image hosting!?" (cough cough Getty, Newgrounds, Reddit, etc)

Hopefully all of us existing artists and "prompt builders" can learn to grow together and coexist. Whether its a tool for brainstorming, a medium to interpret your mental state, or a fun no-effort pasttime, there's no denying this technology is amazing. What I fear the most are riled up artists wanting to place down severe limits, demand permission rights, and even get enough people to make these services taboo/hard to access. Even though this tech has the potential of being the foundation of something much greater in the future, people still say this has the same level of morality, human intent, and troublesome ethical practices of those crappy NFTs.

If its a person's recreation, no.
But an actual artist can still use it to help in their creative process, as this tool reduces time brainstorming, but doesn't invalidate the skills you have with your tools.
Not everyone who can make art are artists, and just because this isn't an artist's product doesn't mean its malicious or bad.

To anyone that wants to get rid of this software or make it a pain in the ass to use, please still know

We will never not need manmade art ever again. It will never get tiring, and since ideas will never run out, manmade illustrations will last for years to come.
It may take some time, but AI art can be separately showcased in a peaceful manner. Completely wiping it out is not the solution though.
Its impossible to make concept art for a story with only AI since it is currently impossible to have multiple correctly detailed characters in a stylized room together, all in the same style and consistent in different angles. It will still take a skilled artist time to finish a coherent piece, just not too much.
You and I may have our own stances on manmade art vs machine art, but we cannot discourage everyone from this "low-effort" medium because it doesn't display any real human emotion. No one should be judging others based on if they have a high effort hobby or not, as not everyone has as much time or energy. Also, if a few, lets say 20%, of all people aren't nitpicky and worried that the protraits, landscapes, and concept art around them being made by a mass-producing, image REFERENCING (not stealing) Stable Diffusion, why should we be wishing to limit our tastes in this way?

Intent, validitly, coherency, and trial/error may be a valid point, but the rest of you all acting pretty unusual bringing up "ethics", "copyright", and "consent" into the topic, as if an AI can take your work directly.
These models are trained off of billions of images across the internet with descriptions, so unless you have a hundred or so public works of art with labels, a company cannot rip off your art style specifically with AI. It doesn't make sense trying to protect other art styles either because you aren't really innovating the world by protecting and replicating what has been done in the past. I would much rather see everyone's take on art compared to someone trying to replicate a Picasso painting or the Mona Lisa.
Also, for all you guys worried about concept art getting ruined, if you try writing a prompt of more than one character preforming an action like (blonde man in a white hoodie reaching his hand towards a pink alien full of orange slime in the desert), you cant even see an alien. All the best text to inage AI can master only a few concepts at a time, but even if you can get your OC walking in a barebones city for example, there is no guarentee the AI can frame your OC at the right size or angle the entire setting you imagined. Humans artists make much more believable shots.
And finally, even if the common family can generate their own vibrant paintings, or an overworked individual can easily conceptualize with zero skill or talent, why try to fight it? You or I can't predict or dictate how human we want our 3D renders, book covers, and decals to be. If somethings a tad sexist or racist or xenophobic, it can easily be called out and denounced. There not as many negatives to these programs as people want us to believe.

Oh my god, why are these disgruntled artists gathering together to protect their own livelihoods even when it can be improved and available for anyone? We engage and improve in all sorts of disciplines to make life easier, safer, and more innovative IN OUR OWN VISION, not to present a product to be marvelled for how human and moral it is. If we stop EVERY new product and object because it has failed a "human heart and soul" test, or it can be made by human hands instead of robots, we would never be able to be innovate as fast as we truly can.
Factories and mega cooperations turning everything into a profitable business that cannot be replicated is bad, but so is gatekeeping passionate people who have less time, energy, or money. It's like having a low opinion on the printing press because it messes with the industry and flow of scribes writing everything by hand, ignoring the social effects and potential it has on society as a whole, and allowing more people (including those very scribes) to write on their own.

Every, and I mean every artist saying "Ah so scarry ai is taking our jobs and ruining our craft" is shooting themselves on the foot by discouraging such a useful tool conceptualizing even more new man-made art from coming out. "Ah my styles are gonna die, you stealing my medium!" Seriously guys, just because you can mass manufacture and use machines for furniture, buildings, clothes, food, and cars doesn't mean your vision is gone for good.

Every, and I mean every artist saying "ah so scarry ai is taking our jobs and ruining our craft" is shooting themselves on the foot by discouraging such a useful tool conceptualizing even more new man-made art from coming out.

Its really revealing when people skew their "concerns" of ai art by saying that it will eliminate the need of artists. Even with the hopeful chances of more lifelike and coherent realism/art, our world will always need fresh new artists and experiences to photograph.
Also, why would you want to argue against political work, prominent figures, and NSFW work? If you find shady behavior, you cant just shit on the software.

Even if we come across a bad person using an evil model for evil purposes, you punish the person not the company
You cant arrest a knife manufacturer when a mass stabbing has occured, you arrest and condemn the criminal.
You don't publicize the video games a school shooter plays, you publicize the external factors that led to the moment. People are getting angry for no reason.
Anyways, thanks alot for

https://youtu.be/GVsUOuSjvcg?t=494

idk why I put this here.

-

⎗
✓
**still wip**
17.79
21.84
24.93
25.96
101
28.2022 108
29.2022 117
1. 2022134
3.2022143
4.2022 147
8.2022 158
11.2022173
13.2022 181
15.2022 195
18.2022 203
212
23.224
26.240
249
9.2022!274
15.2022!288
9.2023 360 whatdahell
17.2023 368
22.2023 380
28.2023 394
1.2023 403
6.2023 429
10.2023 440
11.2023 448 -> 2.18 485 ->2.21 495
23.2023 505
25.2023 525
7.2023 537 -> 3.10 542
13.2023 549
5.2023 594
11.2023 611
19.2023 649
25.2023 663
01.2023 675
15.2023 709
23.2023 800
11.2023 811
22.2923 863
24.2023 923
6.2024 1019
26.2924 1232
If there are >1234, then someone AMONGUS