#🍥|anime
1 messages · Page 17 of 1
thx
most versatile model ever
Make ancient greek stuff but in space 
i read that as "ancient greek butt stuff..."
there are ancient greek and roman butt sex toys
plays random bullshit redhaired Final Fantasy looking girl and casts Zetaflare 😄
what model is that?
cool, what model?
alfahentai
holokuki
and as always I'm the prompt master
you using the LoRAs?
ah cool
nope
I'm aiming to get my mecha more dystopian looking, so not too "new and shiny"
I also want them small, like only slightly bigger than the pilot
mad scientists at work
i like evangelion style
and here's a space ship engine that fell off
@wispy canopy and you serious about it being a hentai model?
Blue eye women stand in forest looking for eclipse
Defo serious
100%. i made it myself. and it's pretty damn good at hentai.
it was drunk fluke tho. but v1.1 was not
here is a sorcerers tower at the eve of the convergence of elements
kids mecha!
the father of all the children of singularity's echo
Nier boss right there....
imma slay bosses.
futa out of the box, no loras
Hahaha
chillout bro
never!
Gotta go do some sutff, be back later, now have a uhh
i like my life now
Appreciate life more otherwise you'll end up in SAW.
Can you try same settings + prompt on saucity sauce latest version?
mechanical dark elf
aight chief. imma need to see if my model gets decent shit out of that prompt
wtf is saucity sauce
Model that is better than kenshi
a very special boy and his sheep
oh these ones arent from that one, these are from other prompts
im not using kenshi
just look at how happy the boy looks and how happy the sheep he's playing with looks
It's a meme
being chased around on his two remaining legs
Damn...I feel bad. My prompt was "literal garbage" Haha

I know. But could you test it?
With same prompt + settings
PROMPT GOD... it's just lyrics
Listen up, strap in, notify your next of kin You're about to take a ride a little on the blazing side Guzzle down your Red Bull, you're gonna need a bucket full You're watching me accelerate and tear apart the interstate A certified monster, I'm an absolute trip Like Otis Redding, hard to handle, so you better get a grip Super fast, super fly, bonafide wise guy Call the morgue and say goodbye, write your will, it's time to die
| Caffeine, I'm caffeine Caffeine, I'm caffeine I'm a bad dream I'm a rad scene I'm a tad mean But I'm not afraid to take you out (afraid to take you out) Afraid to take you out (afraid to take you out) Afraid to take you out (afraid to take you out) Afraid to take you out
| I'm a cheetah on the plains, I'm the highway star A supersonic princess in a million dollar car Blood on fire pumpin' through my veins Weavin' in and out while I'm boltin' through the lanes I'm hyperdrive, overdrive, hit the gas at 55 Break neck, trainwreck, in my presence, genuflect Track roundin', speed-a-soundin', 'lectrifyin', pulse-poundin' Heart-pumpin', brain-thumpin', watch me get the party jumpin'
| Yo, listen up, I hope you like it hot Grab yourself a mug 'cause I made a fresh pot Come and get a dose of my kickass java Fuelin' your addiction with this thick black lava We can do up an espresso, or the cappuccino way Sure to get your fix, venti or grande Sugar, Splenda, Equal, maybe Sweet'N Low Everybody's happy when they hangin' with Joe, let's go
| Caffeine, I'm caffeine I'm caffeine, yow! Caffeine I'm a bad dream (yeah) I'm a rad scene (yeah) I'm a tad mean (yeah) But I'm not afraid to take you out (afraid to take you out) Afraid to take you out (afraid to take you out) Afraid to take you out (afraid to take you out) Afraid to take you out
}```
Just enough
wink wink
k wait
can it make boobs?
Yes
winky
downloading asap
what version?
I sell my services on making prompts designed for YOU
Coffee overdose xD
model?
Aurora
Plus
thanku
k wait
wait thats your model XD
Saucity sauce is mine yes
also with my lora?(the one used in the pics you saw)
Yes
ok
Fingerssss 
btw here is a new pic by my model( @brittle jay yours will come after this pic )
Suuuuper de duper new to SD/AI stuff in general, but I did a test with SD depth/inpainted meshes to see if I can make some kind of decent storytelling content:
https://www.youtube.com/watch?v=I9bijudw3-I&t=0s
Images for this test were actually Midjourney's Niji 5 cause it was the new hotness that week, but I took those into SD for depth maps and the inpainted meshes, then brought THOSE into Unreal Engine to add VFX and set up the shots. Definitely like the 2.5D anime look for still shots like this
Get ready to witness the rise of a hero in the neon-lit metropolis of Helix, as corruption and darkness threaten to consume the city. In this award-winning Hollywood film, Kai is a young man with nothing to lose, drawn into a rebellion against a corrupt government with sinister plans to open wormholes to alternate dimensions.
As Kai and his com...
@brittle jay also on clipskip 3?
hmmm... that gives me an idea tho... how can i get my model to be the perfect lyrics prompter model
I got this
Well I usually do 2 on my model but 3 is fine
k nice, because clipskip 3 usually works better on my model
Uhm... Your tail!
cool, closer to what it intended
Ketchup special effects
Clip skip 2 I think is better on mine, but it should still be fine on 3. And it'll be a better comparison that way
ill do both on same seeds with my last pic i uploaded
Ok
I'll try this one
Eden - Battle beast
And what level is she?!!? XD
not very high... i think.
1 = my model clipskip 2
2 = @brittle jay model clip skip 3
3 = @brittle jay model clipskip 2
Imma upscale that one tho
bruh
Wtf happened
wtf
She's so magical her face deleted itself.
meanwhile I've made something else too
correct me if i'm wrong, but the more you can keep clipskip to 1 , the better?
What is the pnginfo?
😂 wtf is that bro
i feel clipskip comes with a loss of prompt following
what?
No
Dragon Ball Super Episode 5's animation.
What settings did you use + prompt
wait u can change color theme of the ui?
yes
tell me lies tell me sweet little lies
Man wtf
how
i am confused too, ill adjust the prompt a bit
Can someone else try my model on their prompts + settings?
Idk man prompt seems like very similar to one I would use
time to try it on colab 
not bad
idk, it just keeps getting worse, ill try with a different prompt
high quality
Got this with A3CAEWC
heh
1=my model
2=your model
its better, but yea. am i doing something wrong?
you all getting cool different mechas
@brittle jay
cool one
Idk anymore😭
I should start selling my services as a prompt crafter
I wish I had creativity for shit like that
I basically just recycle 99% of my prompts
Is my model that bad?
me too
idk maybe my prompting is weird🤷♂️
What model?
Maybe my prompting weird
Saucity sauce
@cyan burrow made a few tweaks 🙂
I sit down and write an entirely new one, then I modify it until it does what I want. The I start removing things until removing it impacts it
ill try another prompt
still, fucking cool style you got going ❤️
Alright Eface! Who you got that's gonna fight this?!
Maybe my last version weird
aye, people can mod it all they like after
Idk why but I always search the watermark/signature that it generates
"he scoffs: 'Seriously. is that all?'"
lol
1=my model
2= yours
this time its way better
Hmm
and i'm not sure man... i don't think i'd personally want face either your abomination or those cryonicus mechs
I wouldn't wanna fight any of these....
Im going to bed again, now this woman leans in to kiss you good night
Does your model have built in noise offset?
pacific rim vibes
nope
I see
idk anymore bro
well the mechs are going to bed, som imma come up with something to fight abominations!
Man I so wanna be on my PC already and continue working on my game project
obviously, i send in the waifus.
your visual novel?
Yeye
when will it come out?
It's only in beginning stages. I want it to also procedurally generate character reactions and background + tts
Also I still need to optimize my prompt (give more examples to the AI) and perhaps switch to another llm
Though I truly love pyg6B
i am going to die, these guys on civitai bro.. I CANT UPLOAD ANYTHING MAN
Fr. Relatable
why not train it?
Haven't thought about that
its easier than you think
Though it would take some money to rent GPUs no?
Not counting the middiffusion upscale, here's a good one for a variety of types ranging from romantic to vampiric biting to stabbing to swinging scythes around
sophisticated splash drawing of young woman in the embrace of a tall handsome man, he’s a vampire and bites into her neck in a gentle embrace she slams a metal rod taller than the vampire, a war scythe folds out from the metal rod, the vampires explodes into guts and dust blood red rain and dew fills the air, soft light, dim lighting, street lamp lighting, blood red glow (best quality:1.27), (masterpiece:1.27), (highres:1.20), (digital art:1.20), (grimdark illustration:1.25), (Nu-Goth Art Style:1.28) (best quality:1.27), (masterpiece:1.27), (highres:1.20), (digital art:1.20), (digital illustration:1.20), (anime-art:1.25)
Negative prompt: (bad:1.2), (wrong:1.2), (error:1.2), (low quality:1.2), no quality, bad quality, low resolution, lowres, normal resolution, (disfigured:1.2), (deformed:1.2), bad anatomy, (no detail:1.2), low detail, normal detail, blur, blurry, nonsense, random, (deformation:1.2), (text:1.2), (title:1.3), (description:1.3), (logo:1.3), (watermark:1.3), bad proportions, , jpg artifacts, jpeg artifacts, extra leg, extra arm, extra finger, cropped, cropped head, cropped body, no head, no body, half body, unnatural pose, melted face, spaghetti body, contorted, contortion, siamese twins, conjoined twins, conjoined body parts, big breast, massive breast, major breast, big tits, massive tits, major tits, ecchi, hentai, eroge, nukige, slutty, skimpy, pornographic, erotic, little girl, young girl, loli, lolicon, people, persons
Steps: 30, Sampler: DPM++ 2S a, CFG scale: 7, Seed: 3693727730, Size: 1024x640, Clip skip: 2, ENSD: 31337, Eta: 0.66
nope i can do it with my 8gb vram
And I would need to write a whole dataset
Too hard.
nah
Whole dataset??
i think you only need a couple of pages
not too much
That's like 10000 tokens minimum or something
no
edited it cause I didnt use bad prompt
idk, did it a long time ago
Rentry?
nah, that wasnt a thing back then
it was like back in 2021
i think
Damn
Got 4 elemental waifus lined up. i suppose one of them can do it.
Let me search for it myself then. Onto project ai
this could be interesting for you: https://endlessvn.io/
Someone sent me a very new one where you can fine tune LLMs with 6GB VRAM!
can ya give me the link to it?
Not much testing yet to determine how good it is yet
Go to project ai server
AI textgen channel
cant you just send the link
Ok ok
i am in like 100 servers
Again. Not enough testing done yet
Aight, imma go to sleep. gnight
i dont care, i will test it anyways
Was just about to propose that you test it yourself
not now tho
Obviously
funfact: you can run LLMs of sizes 13b parameters on 6-8 gbs of vram
you just need to add --pre_layer (if you have 6gb choose 25, if you have 8gb choose 30-35), now it should look like this: --pre_layer 35
Why even bother making the names readable
This on oogabooga right?
Got an error while trying to install it last time
Maybe next time I will manage to install it
Anyways, good night for now❤️ already 4AM for me
Noted for when i can actually install without an error of some kind
WaifuLabs img2img?
Morning chat
what model did you use? looks nice.
how do you get images like this?
Copy and paste stuff from #1087493421209485393
Tried something else but the image just breaks
good evening chat
Happy lome to you too
😬
what prompts and model is this?
Model's Cetus-Mix 3.5, I forgot the exact prompt but I copied stuff from #1087493421209485393
Also Ligne Claire Anime LoRA
thanks
goth barbie Ken
Is there a so-bin model yet? I want to capture the Overlord LN art style.
This was the first time I ever added something with inpainting, the flame wasnt there
how i do
Nicely done. Adding things with inpainting is probably the hardest thing to do with it and this one looks quite good.
little witch
thanku
i dont like this one much but here is
i like it!
mfw SD doesn't work like I want it to...
smol
Frog-hello 🙂

@lyric zealot
HAI THERE
@lyric zealot having things look real aint that hard, I still have some nudes from a few months ago

how about your success
gotem
yes
im back btw
did someone else get a DM by oowa5 asking for commissions?
haven't really play around much with realistic stuff tho
maybe one day, need to get your updated model, plus all the embedding you used
i did
bruh
oh im gonna fucking destroy these guys at civit ai

I mean, if you are interested in anime stuff
then Luna released a new model, Vil as well
current opinion is great 
what the HELL is emad doing?
do you have links to lunas model?
well they are from december but for being from then, when it was like 3x worse than now. It kinda worked
yes, here for hf link https://huggingface.co/SweetLuna/Aurora credit @quick hill 
i dont know why ppl dont like the pictures i post hmm
maybe it isnt sexy enough lol
maybe because they are real?
not here
oof civitai down?
why whats wrong with real people

yup again 
its the anime channel
oh ok

Maybe you broke the server! Because too many tried downloading it all at once 
downloading what?
Luna model 
oh yea it looks really promising

every other channel here is ead lol
yea ik

well
the anime is also not as popular as before
Im still waiting for more days to pass before plotting the graph again

i think ppl starting using other discords
why tho?
Civitai try not to die challenge (impossible)
civit is back right..?
sureeee 
Drunk Bassist
same but different
hello hey
heh using visual novel asset, visual novel character, digital concept art, and then descriptor for a character etc really locks you into that look and produces a lot of images that can give you good training material
obviously upscaling and cleaning it up first
your lucky to have stable diffusion to work, me in the other hand it keeps giving me a error saying it cannot launch python
good job man your very lucky you have stable diffusion and i dont cause appareantly python doesent wanna launch
if this is your way of asking for help, you will have a higher success rate if you just ask 😛
yeah. i need help
Why does it say this?
yes ive checked path
did you also check version? did you do a pip install -r requirements.txt and pip install -r requirements_versions.txt ? in the main folder of the webui which I assume is automatic 1111?
I dont think so
well try
do i have to uninstall python now?
and than reinstall it again and do it
no
go into the webui folder
is it automatic?
or what are you running?
yeah but is it automatic 1111?
yes
its automanic1111
ok so go into the main folder, click on the address bar, type cmd, press enter so it opens and then type in what I wrote above
this will put your command prompt in that folder
yes
do i click this on top?
yes
and than i just type cmd right?
then write cmd to replace it alll
And when i do this do i press enter?
yeah
ok i did it
now it should be updating and installing stuff
and some of it should already be there so it does nothing with it
its got red and blue letters and its downloading
okay good
now wait until it's done and then do pip install -r requirements_versions.txt
ok so when this is all done than should python be up to date with stable diffusion
once you've updated all that you start it through webui-user.bat
geforce 3080
ok good then no problems there
Ok i think its done now
ok type pip --version
should i close my current cmd and open a new one?
ok
so when i do this should it work?
I just want it to tell you the version so that you know it recognizes the installation
and also tell me which version
or the entire line
is there a youtube tutorial or a guide for this?
probably but after this you dont need to do more
or shouldnt have to
you click on webui-user.bat
always start through that
My friend has this one called "google colab" what is that if your familiar with it
its stable diffusion but he just says that he uses the google colab one
that's when you run on google, probably cause you dont have the gpu to run it locally
but you do
They use Google's computer to run stable diffusion instead of his own computer
now double click webui-user.bat
ok
bro this is such a mess
i just wanna start new and uninstall it
what happened when you clicked on it?
I would personally just uninstall the current installation of python and install the correct version while checking add to path
whats the issue
It doesn't recognize python it seems
you need patience to deal with software like this too
do you have python installed?
you can't get angry and tilt
What version exactly?
deleting venv could work, then you run webui-users.bat and it will redownload everything
i have python 3.10.6 installed
then delete venv
this will find python again
im a noobie what is venv? i dont even remember installing that
oh i see it
delete the whole folder
ok do i completly uninstall it?
Maybe specify the path in the webui user thingy
I forgot how to do that though
if you're gonna tilt and declare defeat, you're better off using a webservice
its not that im mad its just the fact how dumb this is and how much of a process its gonna take to just use a simple ai generator
you need some degree of patience and willingness to run things repeatedly intoerror after error
it's not simple though
why cant they just make this a EXE?
I think there was like an easy installer
so people dont strugle and make the software work themselfs
its almost like im the one programming this
because it's open source and up to each person to do that themselves
yeah idk
or each developer
i dont understand its to hard i guess
how the hell are you done alreaddy
what do you mean
Like this one
u clicked on the bat and it sayed what?
nvm
You tried for like 10 minutes and gave up already
is there anyone here who knows this issue?
cause im sure someone who actually knows what this error is would kow what to do
can you provide a screenshot of the error
I used this 1click installer, seems to work fine for me https://github.com/EmpireMediaScience/A1111-Web-UI-Installer
no but we were willing to troubleshoot with you, but your defeatist attitude makes us not really want to
https://www.youtube.com/watch?v=VXEyhM3Djqg&t=6s and here is one thing
Stable Diffusion AUTOMATIC1111 got updated multiple times these last few months since my first installation video, and it has received a lot of new features. So in this tutorial I will show you how you can install the most complete and updated version of the stable diffusion text-to-image Ai + GUI on your PC for free. You need to have at least 4...
yeah let me watch this for the 50th time real quick
just to accomplish nothing
and this guys talking 24/7 while im struggling
either you're having a bad day or you don't have the mentality to deal with open source stuff
nubby sounds so much younger than I imagined
Probably same error
is there a way where i can just fully uninstall the whole thing and everything included with this so i can just start brand new and download stuff correctly
Delete folder
then you delete the entire folder, if you want to start over with python. Uninstall it. Make sure you only have one python installation
not multiple
ok, im gonna uninstall all my other previous pythons and my git
so i can just download correctly this time and brand new
i tell you one thing, if you think just deleting everything and starting from scratch will resolve the issues u have with the program, u will be up for a surprise
??
I think it might help
theres a chance of it working if i follow the steps
cause now i have no clue what this error is and what i did
xD but wasnt that the way u ended up in this mess?
?
i ended up with this mess is probably cause i didnt correctly download and follow the steps properly
also dude, I gotta tell you straight up and buntly, Stop whining like you're owed our help and for everything to work without putting in any effort. We've all struggled too and you will encounter problems in the future too that may dwarf this one. If this is your response to it, you're in for a bad time
If you did something wrong that caused the error, then uninstall and fix it
do things in the right order
I would guess it's probably the same error
never seen such a mentality
That's what happens when you sit all day on computer and don't go outside to meet normies lol
jkjk
i guess he deleted the venv folder and started
but what actually happend
he just sayed still doesnt work
lol
but did anything change?
how do we know since all we got is "why dont they just make exe"
maybe I'm just an old fart and take things in more of a stride
but still
we need you to show us the error if you want us to help you resolve it
First time I see "old fart" outside of manga
anyway im done.
u guys mind linking me a tutorial for stable diffusion on how to install it since im gonna install it brand new again
ok im sorry
Man they always have that weird thing between their fingers and mouth
https://www.youtube.com/watch?v=VXEyhM3Djqg&t=6s also here was the install guide
my friend also linked me this
so im gonna watch it
that too
now watch em, be patient. I dunno how old you are, but stop acting like a hormonal teenager or you'll never get anywhere with this
this is a old cigar pipe elegant woman used
its like a filter kinda
I also get that on a ton of images
yo guys, look at this sick geometry dash level i found [epilepsy warning] || https://youtu.be/Ow7nDnZTbDw ||
A very unique level in all aspects, a flashy style with bright and powerful colors, impactful effects and a theme that is bizarre, strange and executed perfectly. The eyes come in different styles and sizes, some look cartoony, some look realistic, but they all have their place in the level, the movements and animations are incredibly smooth, an...
little off topic but yea
I am a teenager myself who has asd, ADHD and dyslexia. Idk what that says about them
what dictionary?
inside joke that @brittle jay probably gets
not sure if you are still struggling, but you can try using this one-click installation
super simple 
Yeah I linked one like that
the syndromes that have random letters
Oh

got the dictionary ;O
They might've missed it so it's alr
Ok I shall now go and actually brush my teeth, eat "breakfast" (it's 3:30PM for me rn) and do homework
ah visual novel generating working good
now I can make characters for my VN
and create a short story for everyone
how do you make perfect standing VN character sprites?
I am also working on a vn project
you mean like the ones here?
no i mean the sprites that are pefect front facing
This guid is a pretty solid starting point. https://huggingface.co/hollowstrawberry/stable-diffusion-guide
@merry tendon ?
once I have enough of these I can make an embed, then I can use the embed (or lora) together with control net
control net to achieve a pose
there are multiple ways to do that, you can use controlnet, with prompting, or charturner or many many more
charturner?
it's an embedding
I mean you could have it face viewer too
I think this is like the fourth link we gave the dude, might be too confusing
ah are there different ones for different degrees then or what?
I gave a video that is the easiest to follow
only played with the first version of charturner
by that dude with the accent
so not sure how it is one the current one
i know with the latest charturner, you can even use openpose to enhance it

tried to add "concern" emote, though too low graphic. I need to add some more quality tags
or upscale
sorry for the late answer, my internet went down, i mean like this, perfect standing every time + COMPLETELY white background
for some scenes I will just use the upper body icons like face or whatever to show their emote
control net I guess and specifying white background or using extension to remove any other?
Tbf, you don't even need that good of backgrounds if you want just the character sprites since we have extremely good background removers
fr? never used them
You should try
can ya link me to one?
No cause I forgot where it is or what it's name was
photoshop and clip studio paint, krita etc also have great background removers
I could try to search it again
they can recognize characters really well now
Even better
Than them
I have more control in photoshop
can do a regular bg removal in AI
then a cleanup in photoshop
i am poor, i dont have photoshop
then get Krita
(you can also change settings so it does it even better/ more optimized)
they appeaer confused as to what to do
There also might be an extension for that in a1111
I suck cause I forgot how to make controlnet work for prompts
very cool 
that is a real pic, i asked because if people like it i will train a lora on that style
ok, i will make one then. can i ping you when its done?
Sure! Please let me know
nice
I will try and use it in one of my arts
you can check out my other stuff in the meantime(ive made some "abstract" loras) https://civitai.com/user/ikaridev
idk what to say, never happend to me 😊
I have used it lora a bit and had a great results with it
I used it in this piece (I didn't post it on my socials yet....)
But I really liked the results
its a bit old tho, if you like Darkstyle you can check out my alexchow style lora, it may look cool on your pics
its a bit experimental tho, so appreciate any feedback
no prob! if something goes wrong let me know
btw
Is there a possibility that you might help me with learning how to create my own loras?
I want to get started but I am a bit overwhelmed with all of the tutorials and stuff
yea its pretty easy when you get the hang of it
when im done training my current lora ill show you
Something isnt right..
can you say the full name?
Geometry Dash
cant play it, my reaction skills are so bad. cant even finish the 2nd level
yes
@nova remnant
ill dm you when its done
should i send you the early versions?(while training it creates 4 files)
that would be great!
How does one train LORAs for models like AOM3A3? Basically no dataset exists for that kinda artstyle specific to a character
I tried training on just NAI and then reusing that lora, but that turned out horrible
Fucked something up internally i guess
I tried training the lora on anythingv3 aswell
and on AOM3A3 directly too
Both result in 2d outputs
i as an example train on my own model, maybe your options are fucked up
they did have quite a bad day yesterday on civit, regarding their database. I have multiple of my models that I can't find on there right now without using the direct link, they don't show in searchs
so it also could be a bug
i can see all my models
I documented more precisely for one of my models there. I'm waiting for any response yet :p
https://discord.com/channels/1037799583784370196/1095686262947590256
it says that i dont have access to that link
on civitAI discord
it's a bug report I made there, it should be public but you need to be there
for short
why does civitai have so much problems lately?
they push too much new features without any staging area to really test I think
Got too excited on releasing new features
nope, I'm a community member that spends too much time here and that moderates on his own time
volunteer work if you want
this was just my personal opinion on what's happening recently on civitAI, it's so much more down than up recently
but I'm not reflecting any official position of StabilityAI on such things
I don't think they have any position in that regard
Lately?
Always been that way
🧏🏻♀️

True, Civit do be scuffed
a friends new model version is awesome owo
managed to push the PoW models in 2 versions on civit before even one is done uploading on HF though, right now
Oh, Sis @quick hill was complaining when she uploaded hers that HF was faster 
may change on a per day basis ^^ not sure. it's been the same last week for me
at least one did the job already
civitai down like crazy lately
Maybe, that could be possible 
to be fair, I send the diffusers to HF too, not just the ckpt/safetensors, but still
days ago server have a lot of problem, very slow
yesterday*
probably related to that
yeah, it was mostly out most of the day for me
they had database problems from their alerts on their discord
I wish we had diffusers 
and upload are not through for me
what do you mean ? Who doesn't share those at all ?
I had a bunch of issues with uploading my LoRAs at a time, was like really laggy and really scuffed
you can always build some from a safetensor, but there is quite a loss in quality
My LoRAs too ages to load
And I don't know if they ever made it to the front page
Some people said they did though
I got 43GB to upload each time I do a community model, in total x) takes around a full day
I mean like the ability to load em in A1111 
Oh damn das a lot 
lately, I even get some of my models not indexed, not in search results, just available through direct link...
Yeah, I mean
For my uhhh Aroma LoCon, it like
Got published, but without any pics or any files or anything
Somehow
Like half the stuff was missing, and it was like really weird
I vibe with the uhhh, tagging thing they have on pics, I think thats cool for people that dont want to see A or B, so they can pick what they see, but I mean, its not the best, and I think it all went downhill from when they implemented that xD
I think they should bring back being able to mark pics as +18 yourself xD
inspired by oobabooga
thx
I think horror 
Maybe, either horror or creatures 
XD
non-upscaled 😦 (girl), in an sensual pose, on a mountain cliff, with breathtaking views, surrounded by wildflowers, with wind blowing through her hair, sensual and alluring. petals in the air
really good = 0
It looks good! Can I ask for the name of it?
i want to know too. is it public?
it is publicly available in a beta version on dreamstudio or on the api
The final model isn't out yet to download for personal use like the others for now, nor even finished finetuning it seems
Wha?
Sleepy Guizmus?
I think they were asking about @frozen olive 's anime model that a friend of them made 
You're goods xD
I read SDXL everywhere, my brain jumped on default position
I'm sure you'll have every anime model soon on DS API though (if only :p)
Yeee I assumed that XD
Would be pretty neat dream tho xD
Funky pixel thing 
is there a model that is more background oriented than character ?
yes, plenty
do tell pls
my model is a bit background and bit character.
idk just search on civitai
as someone that doesn't need to be concerned about my opinion reflecting StabilityAI. I think CivitAI are too busy trying to figure out how to make money that they've neglected figuring out how to deliver a good and improving product
keep all your users happy, not just the paying ones
nice
even freaks want to bathe
what is interogate deep booru?
it analyzes the image you have in img2img and fills prompt window with tags like in booru sites
same as here https://huggingface.co/spaces/hysts/DeepDanbooru
i see
do you have any idea i can adjust an image i already created? Do I goto img2img?
There there ototo headpat
could always inpaint and mark her eyes and write a prompt
sometimes i get this problem with img2img, i usually hires the image, but if i send it to img2img, the quality seems so bad after.
use multidiffusion instead
There we go. Control net helped me make what I wanted with the long coat, sleeveless arms, black undershirt and sword
Something I like doing is taking the pic into a photo editor and making the changes myself very roughly, then taking that into img2img, masking my changes and using script to test different settings
You can just copy red eyes from something or paint in your own and then have img2img turn that into something believable
@hearty heart @pliant epoch it is public https://huggingface.co/bluepen5805/blue_pencil 
Ty, I'll have a look and play around with it

https://youtu.be/QtxW39OQbbc Just saw this, I assume this is like american law stuff? :o
😳 16,000 STUDENTS SALE!! Get a MEGA 30% OFF ($174) the ART School for Digital Artists program 🎓 http://cbr.sh/q8zql0 until April 30th 2023 ONLY!!
Join our epic art community! WE JUST REACHED 16,000 ENROLLED STUDENTS! Nani?!
Updated Copyright Office policy: https://www.federalregister.gov/documents/2023/03/16/2023-05321/copyright-registrati...
her outfit is easy to generate i love it
reee face
Inpaint time 
mmmmmm good excuse to learn it yis
Yeah but. Im dum 
funny that they say things AI did do not count as human work and cannot be given copyright. Does that mean that content aware fills and such, especially when used in movies. Would not be copyrighted?

I made that moomer pic 
Ayyyy 
it looks so nice
It do, made more with better hands but they're nsfw
SLE, mksks style, best quality, masterpiece, detailed background, highly detailed, intricate, detailed face, long eyelashes, detailed eyes, (standing:1.2), (thick thighs:1.0), (medium breasts:1.0), (winter:1.3),
(one girl, 1girl:1.1), (solo:1.3), (nanashi mumei:1.0), (brown eyes:1.0), (brown hair:1.2), (very long hair, multicolored hair, long ponytail, feather hair ornament, ahoge:1.1), (choker:1.1), (thick lips:1.3), (eyeshadow:1.3),
(black dress, elgant dress:1.3),
(light smile:1.3), (3d, real life:1.4)
Fun police may be called
Uhhh, not sure
xd
@acoustic agate Lord Trexdel, bad or no bad? :O
XD
Trexdel is afk probs I think :O
he went to bedge 
Ohhh, fair then :o
Though also you can self report, and they remove it if it is, it worked for me 
O TREXDEL WOKE

doesn't look to be too bad, except that the shorts are basically missing, but underneath is low poly and nothing visible
Lord Trexdel has spoken 
yes
Remov 
Is against the rules to use censorship doing bars and stuff 
I recognize this prompt 👀
oh really
wait she is?

wtffffffffffff







