#🍥|anime
1 messages · Page 113 of 1
Nope
just pure prompt and a bit of luck?
i'd probably need luck -> but that's more on me
I fixed it 
boba addicts!
Yes lots of
addicts
that cloud is still sus...
Fixed
We need couple more people and it will be perfect masterpiece
and my poor gpu chugging along 😮
damn, i finally got to the part where i mixed darkalfa v1.1
that model is actually so intelligent for a 1.5 model 😮
dis evil overlord will not be slain
she'll probably be more pacified with just headpats
that's my cats
deployed a multitude of boxes today, the result was as expected
operation "kitty capture" was a success
Just a smol boba
relative boba size keeps increasing
how big can you get the boba cup vs your boba victims
"i hope he didn't notice me"
Fixed.
aight. 12 am. i probably should stop for today. lets do 8 gens on the new master prompt (16k tokens)
Sleep is for the weak
I BROKE DYNAMIC PROMPTS 😮
it seems it doesn like 16K+ prompts
uhh... yeah
i should just make it a wildcard file at this point
i said i was gonna quit, but now i have to test run that wildcard, of course
Patchouli Knowledge is there.
is this a lora?
Man you don't even know how popular touhou is in japan
Literally every single store here even in the most far in the countryside sell something with touhou on it
yes. look a little lower in the text there
very far from Japan. Why is this project so popular?
Idk
Oh wait I think I do
It was a good game for its time (the first ones) and had a very good copyright thingy that made it very easily accessible for fans to make fan creations based on games
And that snowballed
like the popularity of Minecraft from Mahjong? thanks to the ability to create modifications..
i dont get how SD rendered that image with text on bottom left
maybe i should increase the weight from 1.1 to 1.3 for watermark
viola
well, still he tried and he had to sign his work
The pose and/or character? No, just prompting.
The style, yes, but the lora isn't entirely necessary. https://civitai.com/models/110244/drawing
by adding vibrant colors to prompt with a weight 1.3 this came out surprisingly vibrant
1.3 wasn't good enough, but this ((text, logo, title, signature, username)) is working great
I think in minecrafts case it's different
Its popularity didn't originate or mainly come from this
It just added to its popularity
I wanna hold them and move my hands back and forth
Not to be sus
Or anythin

ya got some good alternatives for online stable diffusion usage similar to nocrypt in google colab?
impressive, the only real giveaway that it's AI is the arch
Cute
Reminds me of 2b
Some of my 2b gens get hair exactly that color :b

I have to make elf 2b now
I made her blonde the other day
Probably a heresy in the world of anime
my current model can't do this 😭
im using a lora for this effect
if you are not using sdxl
oh. that's cheating then 😛
then
really lovely ❤️
it's all fair game 
is it just for duality stuff?
^^
seems monochrome, that's a bummer 😦
I mean, you could also use Latent Couple
is it ? 
well true
can you make it do other variations than black/white?
is it better to use the region control in controlnet than latent couple tho?
not sure, im just prompting about black and white tho, since im doing a "devil vs angle" theme as of now
Theres region control in controlnet?
this?! 😮
I don't see it here
i thought only latent couple or region prompter
what's with that pose tho 😮
I did see this in img2img/inpaint tab tho
yup, i saw region prompter below my controlnet
OH wait nvm

Lemme change the prompt and show
I did Flandre and Remilia, I tried to do Asuka and Shinji first, would have been funnier, but the AI didn't wanna do em @wispy canopy
latent coupler or region prompter to get them in their respective spots?
Latent Couple :o
- 2 tools
One for the stickmen
And another for doing the Latent Couple divisions
such good hands 😮

miku
ugh I feel like crud this morning....
joins the art train
does anyone use google colab?
if you do,
to download a VAE would the code be (civitai):
!wget -c 'enter download link' -O ./models/vae/'enter download link name'
i am using it on comfyui for stable diffusion as well btw
sorry i don't use colab, so can't be sure. If you don't get your answer here, you might wanna try at #🤝|tech-support
thank you
metadata included
she is a genius harvester
ok last one for a bit and work. sigh.
This somewhat looks FAMILIAR
from one of yer pic's
maybe, i've generated that OC for a long time
so yeah 
Thats beautiful now make a tomboy
Man I fr already "broke" my steamdeck. Just wanted to install a certain package and for that I had to update my package manager (pacman) so I did it, but turns out I didn't update the dependencies so now it wouldn't work, then stuff happened and now I can't use sudo/su anymore. I tried to factory reset but it gives error (probably because I can't sudo/su).
Anyways now I gotta wait till I am home (like 12 days) to fix this (quite easily actually) with a USB stick.
Man I fr Speedrun breaking my machines
Like I once soft bricked my phone, then just a bit ago I got my partitions corrupted on my PC and now this

Man beanies are best
Man I am fr gonna cry
Like I just bought this like 2 days ago
And this is right after breaking my pc
What's her name?

grown up version of Vil's OC, so not sure 
Hehe
tbh, looks just like any other AI art tho, but i liked it 
What's her name?
I mean
Everything here looks like just ai art
Literally a vtuber design
true 
This kinda looks like a webtoon style
Which lora is causing thay
I mean like the face looks like Korean style drawing something something
because im using a character lora of a webtoon
tbate
This style looks good
Is there a lora on tbate
I read some parts of it but never took much notice to the artstyle
there's only like 3 different characters lora of that webtoon
the specific one i used is here https://civitai.com/models/98469/tessia-eralith-the-beginning-after-the-end-tbate-lora
No style lora?
don't think so
there might be webtoon lora tho
just in general webtoon
can't link it here since it's super nsfw on the coverpage

I am pretty sure I asked you that before
How is she so fabulous like this?
What is her secret?
no idea, she's born cute
Where she get that gun from?
lootboxes
lootboxes you say
haha
uhh someone isn't happy about their loot
like htf
here have mine!
We love her hair, and her expressions.
what checkpoint and/or lora do yuo use for these?
here you go with metadata
thank you bro
i just drag into my workflow correct?
yes, or any metadata viewer
or this website https://supagruen.github.io/StableDiffusion-CheatSheet/
thank you so much

Is this Tess from the manhwa Beginning After the End?
yup yup
changes models, adds some randomness of style pile :D
sets up the art shotgun
oh and using the new restart model 😄
ok. then
dials it back
"get in the mecha shinji"
maybe you make Jasmine too?)
i literally was trying to make Jasmine, but couldn't find any lora for her, and found tess instead
(
beginning after the end so good tho, can't wait for new season!
one of the better isekais
uhh... might be better to keep this one chained
This LoRA: https://civitai.com/models/110244/drawing
On this model: https://civitai.com/models/100505/kakigori
Prompts are in PNG info.
anyone using any SDXL models that can follow prompt to generate full body photos?
yesh
@supple raptor can that style do backgrounds too ?
sort of
Could probably weight the background more to get more of it.

Is it normal for images to look kinda of blurry like that ( on the face specially ) in 512x768 resolution
Yeah, that's fairly average. Eyes could be a little better with some checkpoints, but not a huge amount. Adetailer is made to fix that problem.
An important thing to understand is that SD operates in latent space. This means that each 8x8 pixel block of the image is represented by a single number which then has to be translated to 64 pixels by the VAE. This means that small details are very difficult for it to recreate, the solution is making those smaller details larger by increasing the resolution of the image.
try with unipc thats faster
Toyed around with it some more the other day to make it do a lora alone its niice
@finite herald
tiled vae?
Not hires directly. But you can just send to img2img and use either tiled diffusion upscaling or SD Upscale.
tiled VAE goes a long way towards reducing VRAM usage though
That doesn't work for me
😔
Says it requires no-half
it definitely doesn't require no-half
--listen --xformers --enable-insecure-extension-access --api these are my launch arguments and I use it on all of my gens.
I'll see it again I might be confusing it with something
@supple raptor
do I change those numbers and try again ? >.>
I dunno what this means
can you send the actual error from the cmd window?
Oki one minute
Pastebin.com is the number one paste tool since 2002. Pastebin is a website where you can store text online for a set period of time.
Oh, I bet its a limitation of directml. You're on AMD aren't you?
Yes
Do you have any experience using WSL?
I don't even know what that means 
That's a giant moon
Windows Subsystem for Linux. It allows you to run the Linux Kernel from inside Windows, sort of like a virtual machine. If you were to use it, then you could use standard Auto1111 WebUI with the CMD window being in your linux terminal but still using your browser of choice in Windows to interact with the actual WebUI and without all having to deal with directml.
Here is a guide for installing WSL2 and Ubuntu if you would like to look more into it. You would not need to do step 5 in this guide as a GUI is not necessary for the way you would be using it. https://ubuntu.com/tutorials/install-ubuntu-on-wsl2-on-windows-11-with-gui-support#1-overview
Shiki Tohno, is that you?
Would that make me run the SD as if I'm running it in Linux? @supple raptor I heard it was better on Linux
Yep, exactly. It wouldn't be worth the extra little bit of trouble on an Nvidia card, but on AMD, it's would be a much better experience.
And then how do I install the SD inside the Linux

From the linux terminal, you would just run the webui.sh shell script. You'll want to create a fresh install clone of WebUI from the main A1111 repo and do it from there.
the rng aspect of the SD is pretty interesting following a pattern of a given prompt
This one: #🍥|anime message
is the lyrics from Pictures of You by The Cure
This one: #🍥|anime message
is the lyrics from Time After Time by Cyndi Lauper
This one: #🍥|anime message
is the lyrics from Ordinary World by Duran Duran
Me when I first discovered lyrics prompting
Did you do that song as well?
I've been lyrics prompting off and on for a bit.
Iris by The Goo Goo Dolls
Not these songs
But yeah
No Access...doesn't show which channel.
meh
lol...did Don't Worry, Be Happy really give you an empty bed?
This was very long ago
Just a bit after the nai leak came out
So models weren't the best
That was on nai
La Isla Bonita by Madonna
great
The Final Countdown by Europe
i didnt steal ther other one
what did you do with it?
hid it. now she has to find it
ahh didnt you want to play with it tho?
lol not really but funny you say that
reminds me of halloween
the game i play has amazing halloween festivity with cool decorations every year
dats Marisa Kirisame from Touhou Project
idk who that is.... a mod?

no, character from the games
think i read about touhour project earlier on the web
oh
someone was posting a link before
there is also the Touhou Project AI server (where the Waifu Diffusion models come from) so I can understand the confusion
It isn't good so far. The current one they are training based on SDXL looks kinda promising, so hopefully it will pan out. When I started on SD, Waifu Diffusion 1.2 was the only anime model that was even remotely decent.
do you have the link for that mod i can download and try it out?
WD1.2 had it's own special charm. At the time it was pretty mindblowing, but to put things into perspective, I ran batches of 100-250 images at a time and was pretty happy when I got more than 1-2 that were as good as this one.
There's a preview of their XL model from a couple months ago here. The more recent ones that the people working on training it have shared images from look significantly better. https://huggingface.co/hakurei/waifu-diffusion-xl/tree/main
Yeah that one is WDXL, not the ancient WD1.2 model that I was talking about before. Both are fully finetuned models based (mostly) on danbooru datasets. The one I linked is based on SDXL, hence the big size. They only uploaded the full precision version and you would need to convert it to bf16 or fp16 to even load it on most consumer GPUs.
interesting, but i dont have much space on my ssd .. i had to delete a few models until i get more storage i'd have use selective mods
I might have lied.... I gave the horn I stole to Shuten.
tried the devils horn .. didnt work
cute ❤️
if you are using any anime model, then "oni horns" might work 

Bam ...
halo didn't work ..i mean i see a glow but not quite ..
for me, i put "holy halo"
lol .. ok
must be the model im using
this is what i got with holy halo, but didnt use weight
then it's probably model 
yes most likely, its anime themed but not strictly, the author gave it a 2.5d feel
...I give up trying to find appropriate channel for these.
Donald the tank engine
Train station ads be giving me flashbacks to metamorphesis
i found a touhoe moba game
https://store.steampowered.com/app/878380/Touhou_Big_Big_Battle/ its kinda silly
Touhou Big Big Battle is a Touhou Project fangame, and a casual hobbyist project made by MyACG Studio. We hope you enjoy it! []Train your army, explore a variety of skills, and push down your enemies![]Control your protagonist, hoard the points, storm the battlefield with your spell cards!Develop your unique deck and strategy, crush your opponen...
1180
D:\automatic111\webui\extensions\sd-webui-controlnet\models
vs
D:\automatic111\webui\models\ControlNet
is there a difference. my models are in the extensions dir but wondering why theres a folder in the models folder
IIRC originally they had to go in the extension's directory, but it was later changed to accept either location. As long as the extension is finding them, it doesn't matter which place they're in.
gotchit
@supple raptor
hey
I got an error and the thing told me to go to this site
Do I ignore this nested virtualization part ? >.>
If possible, you'll want to enable hyper-v. It may need to be enabled in BIOS. How to do that exactly will depend on your motherboard. If it's old enough it may not be available.
its a b450 I don't think its too old >.>
Yeah go into BIOS and sift through the settings to find Virtualization or Hyper-V/Hypervisor and turn it on.
okay I'll be back in a minute, have this secret booba technique meanwhile
Can you do that but with vil's OC?
let me try 
this is the closest
I mean this is pretty accurate
https://www.youtube.com/watch?v=EtGDGCxq6m8 this is the best version of that song 
原曲 : https://www.youtube.com/watch?v=D6DVTLvOupE
原AAAAAAA : https://youtu.be/65pJPhP1wzw?t=61
edit. 5/18更新字幕
edit2. 5/25 搬運ok,但請標明出處
What's that
Discord server

I hear darude sandstorm
When I get home I'mma go right back on the grind
🌪️
Though I still have school works and stuff😭😭
you didn't leave your computer on with sd server setup? 
I'mma send you invite to projectai
@supple raptor okay I enabled the thing, seems to work
Used the time to get my ram speed up because it was at 2100 still, it's a 3600 kit
Now the thing is asking me for a Linux username of sorts 
You on Linux?
W
can i ask for help ?
Yes
I'm trying to install some inside-windows Linux stuff
To run the AI as if I was running on Linux
Wsl?

help me with instalation ? i got to Step 5: Setup the Web-UI on this tutorial - https://www.datacamp.com/tutorial/how-to-run-stable-diffusion
Explore generative AI with our introductory tutorial on Stable Diffusion. Learn how to run the deep learning model online and locally to generate detailed images.
Yeah
Why?
what about step 5 do you not know? 
Some stuff doesn't work I'll see if wsl makes big difference
doesnt want to instal it
This tutorial is so bad lol it didn't even link the repo
Anyways
so i thouht it was a falty download
so you tried running the webui-user.bat and it didn't work?
Install what
well yeah, it's inside one of the command but yeah
well i have to repeat step 4 because i thought file was either corupted or outdated
yep, just set whatever username you want to use for the linux install
Oh yeah
hmm I think its done
Just do git clone https://github.com/AUTOMATIC1111/stable-diffusion-webui.git
It welcomed me to ubuntu \o/
in git hub dextop aplication?
Then if something goes bad paste error here
cmd?
C:\Users\stoja>git clone https://github.com/AUTOMATIC1111/stable-diffusion-webui.git
Cloning into 'stable-diffusion-webui'...
remote: Enumerating objects: 27343, done.
remote: Total 27343 (delta 0), reused 0 (delta 0), pack-reused 27343
Receiving objects: 100% (27343/27343), 32.24 MiB | 20.54 MiB/s, done.
Resolving deltas: 100% (19167/19167), done.
do i need to paste it somewhere ?
@supple raptor should I do this too ?
In the C:\Users\stoja
where is it located ?
Just sent
Should be named stable diffusion something something
Yes
C:\Users\stoja>webui-user.bat
'webui-user.bat' is not recognized as an internal or external command,
operable program or batch file.
C:\Users\stoja>
nubby right now 
Technically you can do it through cmd
But you're not in the stable diffusion folder that's inside the user folder
and copy paste? to intended folder from the online tutorial?
He always like that
He types slow on phone I think
it isn't necessary if you're just using it to run WebUI. systemd is used to automate tasks. you can do things like setting it to run certain processes as soon as you open your new Ubuntu install. i'm assuming you aren't very familiar with linux commandline so I would start without it for now, and you can always add it later.
lmao
nubby is friend he helps 
not even on phone right now. rum made my brain go slow
Hehe
also, i'm old. don't make fun of my slow phone typing
so what should i do?
I could probably type blind on a t9 keyboard as fast as you can on your phone
go to that folder on your window explorer, and run the webui-user.bat
not making fun, just always anticapting what you are gonna type 
i got 2
with great knowledge comes great responsibility
On the right you see their type
You see windows batch file?
That one
use the windows batch file 
Also you might wanna enable the ability to see file extensions in file explorer
Creating venv in directory C:\Users\stoja\stable-diffusion-webui\venv using python "D:\AI i Stable difusion - lora\python\python.exe"
venv "C:\Users\stoja\stable-diffusion-webui\venv\Scripts\Python.exe"
INCOMPATIBLE PYTHON VERSION
This program is tested with 3.10.6 Python, but you have 3.11.5.
If you encounter an error with "RuntimeError: Couldn't install torch." message,
or any other error regarding unsuccessful package (library) installation,
please downgrade (or upgrade) to the latest version of 3.10 Python
and delete current Python and "venv" folder in WebUI's directory.
You can download 3.10 Python from here: https://www.python.org/downloads/release/python-3106/
Alternatively, use a binary release of WebUI: https://github.com/AUTOMATIC1111/stable-diffusion-webui/releases
Use --skip-python-version-check to suppress this warning.
Python 3.11.5 (tags/v3.11.5:cce6ba9, Aug 24 2023, 14:38:34) [MSC v.1936 64 bit (AMD64)]
Version: v1.6.0
Commit hash: 5ef669de080814067961f28357256e8fe27544f4
Installing torch and torchvision
Looking in indexes: https://pypi.org/simple, https://download.pytorch.org/whl/cu118
Collecting torch==2.0.1
Downloading https://download.pytorch.org/whl/cu118/torch-2.0.1%2Bcu118-cp311-cp311-win_amd64.whl (2619.2 MB)
- -------------------------------------- 0.1/2.6 GB 18.7 MB/s eta 0:02:15
Bro
will i have a problem
Wrong python version
with python?
aaaa how do i stop this now?
Maybe you had used python before?
it's installing torch right?
Close cmd
ok closeed
Torch no work with GPU on that python version
That's why you have to use that version
yes, you do need to install python 3.10 (I use 3.10.11 instead of the recommended 3.10.6 which includes some bug fixes and does not cause problems with AI stuffs)
I'm using 3.11.3 💀
technically don't need to uninstall, but it's easier that way.
Im lost

In which part?
how do i uninstall?
Only need to uninstall python yeah
Yeah that makes more sense
The part of: how to install sd inside this linux
Git clone the thing then type in terminal ./webui-user.sh or something like that then voila
Though idk might be a bit different on wsl

I'm hacker now
Also type that inside the SD folder yeah
The one you cloned
With da git
got my python 3.10 now have to add it to PATH
- search for "ubuntu" in the windows search bar and launch that
- use the
CDcommand with the syntaxcd /mnt/DriveLetter/folder...to get to your stable-diffusion-webui folder - use the command
webui-user.shto run the shell script and launch the WebUI. - wait for it to create the venv and install dependencies just like when you launch the webui-user.bat file in windows.
Man I sometimes just feel like:
"X doesn't work the best on Linux and its X's fault but some guys on Linux worked hard and now it works slightly better but it could be easily fixed by X"
Literally python be like
in the cd command, you'll need to enter your own drive letter and folder structure for wherever you have it stored.
Cat has AMD GPU so Linux is their only real option.
It's because WSL has to mount your windows drives in order to access them because the kernel is running inside of windows.
Just start with arch then accidently purposefully delete your boot partition and brick your pc
Or install kubuntu/Ubuntu with gnome and don't even use the terminal
And literally same functionalities as windows but better
I'll just keep my linux isolated to WSL and my Oracle Cloud instances.
C:\Users\stoja>python --version
Python 3.10.11
C:\Users\stoja>
ok got my python 3.10 and added to path should i go again for instlation?
yep
and what happend to the first instal i tryed ? are some files left over ? should i delete them?
Do I do the first steps of installing the SD in windows to create a new stable-diffusion-webui-directml folder, then do that ?
If you have a folder called venv inside your stable-diffusion-webui folder, delete that just to be safe.
Nice man. Like there was a guy like a couple days ago that got stuff in this state because of windows
Fuck windows
Now I imagined something malicious
you can play games on windows
On Linux too
Most games already work well on linux
did you already clone the Automatic1111 stable-diffusion-webui repo?
I didn't do anything yet, I want a new fresh install for this :d
i think i did have
Except games like valorant that use anticheats which can literally do whatever they want with your computer (literally anything. They could ruin your computer in a second if they wanted)
You still have
stoped playing riot games a good 5 years ago
yep, go ahead and delete that venv folder. that's what it's building when you run it for the first time.
And I think also COD just dislikes Linux so they just won't let you play even tho you can
you can leave that
ok
i guess all is good for now ?
Creating venv in directory C:\Users\stoja\stable-diffusion-webui\venv using python "C:\Users\stoja\AppData\Local\Programs\Python\Python310\python.exe"
venv "C:\Users\stoja\stable-diffusion-webui\venv\Scripts\Python.exe"
Python 3.10.11 (tags/v3.10.11:7d4cc5a, Apr 5 2023, 00:38:17) [MSC v.1929 64 bit (AMD64)]
Version: v1.6.0
Commit hash: 5ef669de080814067961f28357256e8fe27544f4
Installing torch and torchvision
Looking in indexes: https://pypi.org/simple, https://download.pytorch.org/whl/cu118
Collecting torch==2.0.1
Downloading https://download.pytorch.org/whl/cu118/torch-2.0.1%2Bcu118-cp310-cp310-win_amd64.whl (2619.1 MB)
--------------------------------------- 0.1/2.6 GB 18.2 MB/s eta 0:02:21
Very good
go ahead and run git --version in that Ubuntu terminal to see if that was included because I don't remember if it is normally or not.
Nice
Now you gotta wait
Don't close this
i know
It says git version 2.34.1
Git is included in most distros by default I think
Except like probably gentoo
lol
cool, now you can cd to the directory you want your Webui folder to be located in and run git clone https://github.com/AUTOMATIC1111/stable-diffusion-webui.git
does someone want to check my progress in pm where i can screen share?
I figured it probably was, but didn't want to just assume and leave them confused if it wasn't.
Makes sense yea
Press the print screen button on your keyboard and then do ctrl+v here
Should send a screenshot
big question how large is instalation ?
ok
I am out of touch
Oh then that should be enough
should have done this on D particion
Yes
left it empty just cuz of that
You could just move the folder I think
Like this ? 
Nubby can you confirm or deny
Type ls
moving folder will break it for git pull. I don't know how to fix. I always just pull a fresh copy if I want it somewhere else.
Good good
What is that for ?
add cd in front of that
Shows you what's inside the folder you in rn
if/ when i start to generate picutres will i have space or maybe will i be able to specify path to D particion?
I am only confusing myself frfr
Uh, you probably should be able to
images are pretty small and you can easily change the output directory in settings if you want.
But models take more space than images
So you should change the models folder
Do that later tho
No need to rush
can i take particion space from D and add it later back to C?
Depends
Are they on different drives
Like physically
1 ssd
Then yeah
if they're the same physical drive, then you technically can, but I wouldn't recommend trying that with your boot drive. things can get fucky
But playing with partitions is traumatizing for me
I got there by doing the cd command, I did the git and it installed the files 
This
i think i am done?
nice now run that webui-user.sh command from inside the new stable-diffusion-webui folder that was created and it will start building your venv
girl playig with legos inside a cavern, lit by candles, eerie smile, looking at viewer, dark, shadows

now that scares me a little - what will be the space ?
and is this done?
my venv takes 7.44GB with quite a few extensions adding to it. I think without any extensions is roughly 5.5GB.
can i close it?
How do I do that command, I'm just typing webui-user.sh in there and doesnt work lmao
Says command not found
type pwd to show us where you at
no, the cmd should be kept opened while you are doing your generation
@lyric zealot
I did the cd thing to go in the new folder and dunno what im doing next 
aaaaa what does that mean ? what happends when i shut down my laptop?
and now ls
when you turn off your laptop, that cmd is closed as well. So after next startup, you need to re-run your webui
im unsure if it's webui-user.sh or webui.sh
should be the first of those.
User

Tho technically both work
try ./webui-user.sh
than can i just copy paste it to particion D_
./ indicates the current directory.
like nubby said, there might be some issues with your git clone later on
Basically with doing updates
yup yup
nothing happened >.>
Either you reinstall the thing (git clone thingy command) on your d drive or just make it use models from d drive. Models take most of the space anyways
maybe my brain is being dumb. how about bash ./webui-user.sh?
It should use bash by default
that reinstaling would be perfect
Why is there no output??
I would think so too, but something is wrong here.
So just like delete the stable diffusion something something folder
idunno
Try the one without user
can you cat webui-user.sh just to see there's stuff inside
Then you need to cmd into your D drive
Google how to do that
I am too lazy to explain
Then do the git clone command you did before (the one from the tutorial)

cmd on linux?
oh
There are 2 people installing the webui simultaneously
I did the one without user
One in wsl and the other in normal windows
and it screamed at me
not playing with lego but
Like this ?
sudo install python3.10-venv
okay
The thingy
pseudo apartment
sudo apt install python3.10-venv
||sudo rm -rf|| don't actually do this
Me when
This is the equivalent to deleting system32
I think
Idk I forgor
Yo
Yo
many things appeared
well, in this case, it's probably just deleting any and all subfolders from the current folder. But if you change it to ||sudo rm -rf /|| then it's gg
sudo apt install python3.10-venv is doing things o:
Now run the sh thingy
was it with bash ./webui-user.sh that you got the warning about python-venv earlier?
it was ./webui.sh
Do that
a question how do i use Lora than ? do i have to download some files ? and how do i train AI to generate better ?
u download lora files and put them in models/lora folder
i do not have Lora folder ?
if u installed correctly you would have it under models/lora
should i create one my self and add files or it needs to be done via code?
you should be able to create the folder yourself
oh no
you can but is your models folder empty?
download the loras you want to use, put them in .\models\lora then access them in the webui by clicking the Lora tab under the negative prompt box. If WebUI is already running when you place them in the folder then you'll need to click the refresh button that appears under when you're on that tab first. Click the tile for the lora you want to use and it will add it to the end of your prompt like <lora:NameOfLora:1> you can change the number to adjust the strength, typically 0.7 is pretty good for most loras, if you use more than one, you'll likely need to lower each of them.
-
How to use lora - download lora, put in right folder, select it in prompt through clicking on a button then it is gonna use the lora.
-
Do I have to download some files - mainly models and LoRAs and embeddings and extensions. Models weigh the most (could weight like minimum 2gb per model and upwards of like 9gb if it is a weird model, most are at 2-7gb.)
-
To train ai you need good GPU and depending on that you need to download another UI (user interface) that will help you with that. What you need is a collection of images and "prompts" associated with them, this you can automate, do practically you just need a collection of images and a good gpu.

It do be like that
Yes if it isn't there
You do have the models folder tho?
Yes?
you gonna make me watch edgerunners for the 6th time 
It's not that good tho
Personally speaking
perfect time before cyberpunk phantom liberty drops

It's pretty good imo 
I didn't even play the base game. I quit after the prologue. I'll probably pick it back up and actually finish it once theres a sale.
I mean it's like 3.5 out of 5
Imo
But I do high ratings usually
actually perfect tbh. Because with the latest dlc, there's gonna be a cyberpunk 2.0 update that fixes adds what was promised
it should have gotten 24 episodes

Would have been better
and what do you rate as your best anime? or top 3?
Monogatari series
yes
do you guys know what I can do after the error cannot activate python venv
i'm a sucker for violin, piano, and sad anime. your lie in april takes all 3 of my top 3 spots.
hidden gem 
wait what happened? 
@lyric zealot this
https://www.youtube.com/watch?v=SnXkhkEvNIM i love this song, but top 3 most played anime song 
★2017年10月のGoose house Streaming Liveは
10/28 20:00(JAPAN TIME)START!
詳しくはhttp://goosehouse.jp
▼Goose house OFFICIAL INFORMATION
[OFFICIAL WEB]http://www.goosehouse.jp
[TWITTER]https://twitter.com/GoosehouseJP
[INSTAGRAM]https://www.instag...
- Monogatari series second season
- Nekomomogatari black+white
- In third place I am actually not sure, probably like neon genesis evangelion or kids on the slope or something
is that the one who is a blue alien and commits genocide against a planet?
No
did you already installed python3
no its the good one
That's the other avatar

I did sudo apt install python3.10-venv and it installed things
I accidentally skipped the intro once and immediately restarted it so I wouldn't miss it
My favorite anime opening https://www.youtube.com/watch?v=0YF8vecQWYs
美波「カワキヲアメク」(TVアニメ「ドメスティックな彼女」オープニングテーマ)
Minami-「Kawaki wo Ameku」(TV anime「Domestic na Kanojo」Opening song)
作詞・作曲 美波
music and words by Minami
vocal by Minami
Animation:UrataAsao
Movie:tissue
Gt. Otsuka Takumi
Ba. Nanbu Kazuma
Dr. Takashimizu Kandai
Pf. Iga Takuro
Str. Muroya Strings
Streaming platforms→https://www.373official.com/pages/501...
try python3 -c 'import venv'
it should not return any errors
love that song, but daymn that anime 💀
I did it, nothing appeared
i don't know how i read the manga of that series when i was younger
try python3 -m venv venv/
i like elfen lied
elfen lied is great
kuzu no honkai anyone? 
damn those are really old
elfen lied was the first anime wher ei was like "oh anime is sweet" i had only watched dragon ball z before that
isnt that the anime about a guy who fucks around but nothing happens
I think it broke
well, i think many guy fuck around in that anime
what happened?



😳