Previously: >>8455765The cartoony side: >>>/aco/csdgNo DALL-E generated stuff.No futa.>>>/aco/futasdgReport trolls.>GuidesGeneral: github.com/AUTOMATIC1111/stable-diffusion-webui/wiki | rentry.org/sdg-link | rentry.org/rentrysdHires Fix: rentry.org/hiresfixjan23Inpainting/Outpainting: rentry.org/drfar | rentry.org/inpainting-guide-SDAnimation: rentry.org/AnimAnonCreate LoRAs, TIs: rentry.org/59xed3 | rentry.org/simplified-embed-trainingList of models/LoRAs:rentry.org/acorepo>Auto1111 extensionsControlNet: github.com/Mikubill/sd-webui-controlnet | rentry.org/dummycontrolnetRegional Prompter: github.com/hako-mikan/sd-webui-regional-prompterTag autocomplete: github.com/DominikDoom/a1111-sd-webui-tagcomplete>Related>>>/b/degen>>>/h/hdg>>>/e/edg>>>/d/ddg>>>/g/sdg>>>/trash/sdg>>>/trash/slop>>>/vg/aids>>>/vt/vtai>>>/y/ydg
Preventive archive dump
Good thing I checked before hitting post on a new one.
>>8469053That's my biggest fearThe dreaded double bake
End of my archive dump
>>8469056With the damned 5 minute waiting period (that doesn´t stop the bots) it is worse.
you really picked that OP image for the cartoon board huh
>>8469108This is the realistic thread
>>8469117i don't see how that makes it any better.
>>8469127Our OP images are always non-comicsReal brain buster, huh?
and a couple in wallpaper aspect - stretches out the torso too often
>>8469169Fine pâtisserie baking, as usual.
haven't had much time to play lately, so I've just been running batches off my old prompts. any better pony realism models lately?
>>8469205Seems like everyone is running around like chicken without their heads due to Flux. It went pretty much the opposite way of SD3.
>>8469325Why isn't there a Flux thread?Is it bad at making lewd stuff?
>>8469452Why should there be, this is a thread for all models. It's censored like every base model, gonna need some finetuning on nsfw first. And the hardware requirements are much figher than sdxl.
>>8469046>Related>/b/degen>advertising the pedo board that can't stop posting realistic children and I2I of real CP
>>8469504based
>>8469452This is itLast thread had a lot of flux gensThere is visible progress in lewdnessSD also started out like that>>8469504That is so the people go there instead of posting their filth here
>>8469513oh for sure that's the reason you added it and put it at the top.
>>8469529They are roughly in alphabetical orderWhat kind of troll are you?
>>8469504It's like having a retarded cousin, you may hate it but you're still related.>>8469046there's a thread on /c/ now as well if you want to add that
>>8469046>>8469585Next baker shall make this the list:>>>/b/degen>>>/c/kdg>>>/d/ddg>>>/e/edg>>>/g/ldg>>>/g/sdg>>>/h/hdg>>>/trash/sdg>>>/trash/slop>>>/vg/aids>>>/vt/vtai>>>/y/ydgAdded /c/kdg and /g/ldg, in fully alphabetic orderLinking a board is no endorsement, it's to point people to the right placesWe missed baking for a few hours and the haters smell blood in the waterNagging about OP image and /b/, what ridiculous bullshit
>>8469728>>8469742Flux makes smoother skin than XL evenOn /g/ they are discussing that it's a weakness of current LoRA training methodsCan it make photo realistic skin texture reliably without LoRA use?
>>8469748Dunno I'm using CyberRealisticPony
>>8469762Sorry, I though you were doing flux and a nude bathing LoRA came out or somethingShould have looked closer>>8469643>>8469648>>8469650These have amazing skin texture, only looks like flux from the thumbnail maybeGood stuff
>>8469769Thanks. I meant to say I also often use Hires. fix upscale using 'DPM++ 2S a' which produces pretty nice skin texture detail.
>>8469509Catbox?
>>8469046man this gives me an idearequesting someone generate a girl or mermaid swimming in cum
anybody have any Dafne Keen lewds?
>>8469046How do I uninstall this? I'm trying to free up space on my computer.
>>8470061Model?
>>8470061based
>>8469635Love this one.>>8470235Fracking is getting a bit out out of hand.
>>8470890>>8471321PonyRealism. I'm using vicineko style with negative weight to make her look less anime in the face.raw gen: https://litter.catbox.moe/t6i1gr.png
>>8469046So am I right in thinking I need;>32GB Nvidia Card>32GB of memory on mobo>linux>FLUX.1>ComfyUI>some plug-in models for NSFW, BJs, etcjust to make my own pics of ANYTHING I can imagine?>tfw I can already use text promptsIs using ComfyAI like using text prompts but each comma-separated part is just another node in the process that gives me far more control, or keeping parts I like and don't want to change, eg;>locking the random seed>keeping faces>keeping clothing>keeping background?I'm interested.Thanks in advance /aco/.Any tips? Or am I completely wrong?I'm an intrigued noob who cannot draw or take good photos but always wanted to make art.
>>8471388Why don't the online FLUX.1 Schnell do nipples?They say they do;>nsfw>nsfw+>nsfw++Is there a way to reactivate nipples via prompts?Anyone?
>Victorian Mermaid and (big full pert breasts) and cleavage and (topless nude) and (big eyes) and (wide eyes) and eyelashes and scared and (looking at viewer) and (wet muddy pale skin) and (long Victorian updo auburn hair) and (lying on beach), 1girl, nsfw+, cinematic, high detail, high definition, HDR, blurred background
>>8471550
>>8471524There are no consumer grade 32GB cardsOn RAM, yes, minimum is 16.Yes you can do pretty much anything, but Flux, is still too recent, great strides but there are proven and well worked models in SD 1.5 and SDXL.ComfyUI is nodes nodes nodes.The seed can be locked. For the rest is harder, changing seed might alter those too much, better use LoRAs.>>8471550Nice.
>>8471564>There are no consumer grade 32GB cardsYou don't need a gfx card with the grunt... a second hand NVidia Tesla with no output ports is all you need.Stop thinking of them as GFX cards and as AI-Accelerator cards instead.
>>8471561You know what sailors are like...
>>8471524>Any tips?Read the guide in OP and gen a few images before asking questions.Anyway most of that is wrong. Flux only needs 24gb and it's imo too new to do much with, current models are fine with 8gb. Windows is fine too.>some plug-in models for NSFW, BJs, etccalled loras, or LOw-Rank Adaptations>my own pics of ANYTHING I can imagine?Anything you imagine but limited by the model's knowledge and prompt understanding. That's the current issue of Flux barely being able to do nudity, not to mention sex. And why we're using SDXL-based ones until a competent finetune comes along.>Is using ComfyAI like using text prompts but each comma-separated part isNo. It's the same as using text prompts through any other UI, only you get to see the moving parts underneath. Though not like having a transparent gearbox in your car, more like having to rebuild the gearbox out of lego bricks every time you want to shift gears. You're better off using Autimatic11 or Forge, only they have yet to implement Flux support.
>>8471571>gen a fewI am.>flux only needsFUTURE PROOFING.>gflux can'tThese are all FLUX.1 SCHNELL prompted online.>ComfyUII am looking forward to having more control after dicking with and fighting the text prompts for long enough.>hw = old Dell server taken from dumpsterI guess the 96GB and 4 x 1TB drives will come in handy after all.You'd be amazed what's cheap right now with so many companies going bankrupt and the glut of 'office equipment'.Will get an NVidia Tesla with 32GB on it next month for about 80 quid in the UK.I have my eye on a few.>it be hardCan't be any harder than anything else I've used for PCs for over the decades.
>>8469048Got any more like this? Love the style
>>8471524Spoiler alert: You cannot make pics of anything you can imagineFirst law of AI: It can only make images of something it has seen before during trainingThink of it as a giant remixer of imagesSo unless you run training yourself and have reference material, there are notable constraintsStill hella fun though, and there are always things to discover in the models>>8471598Was part of the archive dump, taken from a February or March thread
>>8471615It may be that my imagination is limited only by what countless other have already done.If an image of it exists, anywhere on Earth, anywhere on the web, it will be in the system eventually.I have no illusions.Humans are very rarely original and once they are, it is there for all to utilize.>thousands of years of human art from across a globe to draw fromIt'll be fine.
>>8471615>reference materialSo just a scanner is required?Barely an inconvenience then.
>>8471649You can always teach it new things, but that needs existing images
>>8471532Base Flux only does very weird nipples or red splotches.I'm using nudity loras with Flux Dev.
>>8472177I missed the jokeanyway she's supposed to be ten feet tall
>>8472210And he says he's tall too.
>>8472210no joke, just a nude gen, it goes hereFlux is good but still messes up scale often, doubt the lora was trained with multi-subject images showing Na'vi being taller than humans anyway
>>8472177>>8472242>I've heard Pandora's a beautiful place>Well if that were true why can't you people stay there?>We're going to build a space wall and the Navi will pay for it
>>8472871Go back >>>/pol/
I'm prompting for hands on shoulders but he often gets more handsy.
>>8473064On the one hand, first fappable flux gen I seeOn the other hand, shows all the same issues as SDIn fact if I had made this, I would have kept rerolling for better handsThe background has a weird composition too
>>8473141I definitely could clean up but Flux messes up hands and backgrounds much less often than SD. Even with SDXL this would require regional prompting and then the interaction and the variety of gens would suffer from it.
>>8473242"I am always surrounded by beautiful women. They love being around me. Love me. They never ask me to build a wall across their bedroom door" stated Donald.
>>8473064>>8473242SDXL has no problem making these presidential medal shotsCan get away without regional prompter in this particular case since it's hesitant to turn the president into a naked ladyIt's really hard to get a perfect gen with all good hands, toes and faces like your second image thoughSo flux still requires lots of cherry picking, but the higher success rate is a big plus thenBut it's not groundbreaking sadly
>>8473337>>8473340Go back to your /b/ threads.
>>8473407Those sailors seem to be a bit dehydrated...
>>8473323>SDXL has no problem making these presidential medal shotsHadn't done these in a while so might be mixing up 1.5 and XL gens in my head.It's not just less cherry picking for hands, it's also consistency in placing subjects like x to the left, z behind y, etc. I hope big porn finetunes, if they even happen, don't destroy the base abilities of the model.
>>8473407Eye m'hearty. Run rations be low.
>>8473416"Big Tommy be wastin' his seed on the 'other one we done caught in the net. Don't fancy it much meself but Big Tommy is a strange lad..."
>>8473634>rumored to give unbelievable handjobs
>>8473640King's knight to Nipple 5... Check.
>>8473416Some not all. Depends on the rum ration, lad.
I'm going to start cleaning up some DALL3 gens (pic possibly related) using SD inpainting. What is the best Lora for inpainting photorealistic gens?
Is it possible to make a Lora of an AI person?
>>8473917yes
>>8473917You mean a random person you created in a gen?Training from a single image is possible, but painful
>>8473922yes. Do you have a guide?
>>8473927Just https://rentry.org/59xed3#rare-characteroc-howtoI could have sworn there was a line in there about single image trainingMaybe it was in another guide that I can't find right now
need more Kamala
This is from Anteros XXXL, from what I've seen it's the only model which actually knows how to consistently make assholes properly, and doesn't either leave them out or make a 2nd malformed vagina. Are there any other models which can do this?
>>8474006More Anteros XXXL
>>8474006Anteros XXXL
>>8474006this is a turning point for AI
>>8474006Anteros XXXL. Note that it's not that great at making multiple woman images, you have to mess around to get something like this which doesn't have too many arms or mangled parts and so on. But it completely gets the vagina and asshole parts, that it can routinely do.
>>8474190
>>8474191
>>8473323>But it's not groundbreaking sadlyWhat would you consider groundbreaking?
>>8474458
>>8474460
Can we get some busty Emma Watson getting blacked?
>>8474468make it yourself
you guys know the internet is full of porn that doesn't look like an uncanny valley nightmare right?
>>8474243Not fucking up the hands anymore for startersAsk yourself this: It is X times slower than SDXL, but do you need more than X attempts at a perfect gen?I wouldn't mind a model that's 10 times slower even, but only if perfect gens are more than 10 times as commonThat would be groundbraking>>8474514But I can't control what's in that porn myselfAll gens are made to the taste of one person
>>8474006>>8474008>>8474012>>8474017>>8474031are those images from Anteros XXXL
Flux is BASED
And redpilled
>>8474605>>8474606Go back to pol please
>>8474650nta but not sure why that triggers /pol/ for you - I'd have said /x/ desu.
>>8474509I don’t have a pc
>>8474666Excuse me miss, I've stopped you for a hairy pussy violation.
>>8473883>"...Unda da Double-D, Unda a Double-D... down where it's wetter, down where it's better, Unda da Double-D..."
>>8474527>All gens are made to the taste of one personThe word you just described is;>ART
>>8474701>She's looking at a 8 to 9 inch stretch...
Do you know how to get this kind of photo?
>>8474822Photo-shoots at;>your local senior's care-homethen at;>$1m seaside villas that are easy to break intofollowed by;>make a LoRABarely an inconvenience after that.
>>8469170holy shit, i'm in love. you're just baking these all yourself?
>>8474927Very artistic!(Model, prompt, etc... ? ^^ )
>>8474961VendoPonyRealistic model"nude woman with large breasts spitting fire, BREAK, side view"+ all the usual Pony scores in both positive and negative
Any bimbos?
>>8475030
>>8474862Whoa, what loras and checkpoints did you use to achieve this look? She is perfect.
>>8474862>>8474860>>8474859SAVED
>>8475187>>8475154Thanks for liking it, here are my settings https://tensor.art/images/767348031432597204?post_id=767348031428402902
>>8475204
>>8475228>>8475204excellent!
>>8475229Thanks!
>>8475228>>8475240Can you do Brie Larson as captain marvel like this?
>>8475242No, because I don’t want to
>>8475243How long does this usually take?
>>8475255Maybe two minutes at maxed out settings
First question or thing I noticed? (comfyui user.)What am I doing wrong? Im using loras and got it all setup, but without loras on some of these checkpoints like sdxxxl I cant get a dick to appear with some people? By some people I mean some celebs work by default I dont know why but they do with NO loras. But I cant get a dick, or cum,i can barely get them to touch It runs like a censored bing which makes no sense when I look at all the posted nsfw images. How do I get the default people in sdxxxl to do shit, anything? I tried loras and the male just vanishes.Second question:Is it the loras? So an example with Dreamshaper XL is Big Trouble, its a set. it looks great but if you add a dick or jizz or anything it loses its mind, or at least doesnt know what to do. Do I need some kind of unlocker or what to make them more nsfw on dreamshaper? Are these people releasing these loras with some kind of code setting that disables nsfw or something? I dont see a warning or anything saying its censored.
Does anyone have any idea how this could have been made. I assume image to image but it looks way too polished. requesting a similar result
>>8475243Hit us with another Shota Spiderman one. Makes her so much sexier.
I guess this goes here.
>>8475277Control net probably and maybe a lot of inpainting.
>>8475277controlnet
>>8475305based what kinda specs do I need to run SD I've been using other shit for a while, My computer as of right now is pretty bad.
>>8475332What kind of specs do you have?
>>8475332if you still say stupid shit like "based" i highly doubt you have the intelligence to actually use stable diffusion
where are my chocolate bitches
>>8475845>>8475924>surrounded by washing machines to wash off the chocolateOh AI, you little comedian you.
>>8474969Nice, thank you !
>>8475924Right nipple needs fixing but otherwise great. Which checkpoint was used?
>>8474933yeah, I built a big dynamic prompt with lots of wildcard files for clothes/location/poses... ponyrealism with detailedXL, add-detail-XL, sinfully stylish, and expressiveh loras at 0.4-0.6 and random celeb loras at 0.2. I've also been using CD Tuner to mess with the contrast, brightness, and saturation a little
>>8475971Good question. It was just some pic found over reddit.
>>8476035>tfw where is she hiding the ballpoint pen the time-traveler gave her?
>>8475979These are incredible
>>8474969BREAK doesn't do what you think it does
>>8476062thanks! it's definitely a crafted style though... messing with weights or keywords when you're adding this much detail (mainly for freckles/skin) either burns the image (even at 1.5 CFG) or gives nonsense pixels output. even changing "redhead" to "blonde" gave me major image burn on everything before I lowered some weighting. I think I'm throwing too may loras at it, but removing even one screws up the style.
>>8475845
>>8476080>tfw my bed, after the massacre, had human feet too
Q. What interface are people using?Is ComfyUI worth the effort learning?
>>8476169Russian cops handle arrests slightly differently than those in the usa, it appears.
>>8476175>tfw setting 'WhoAteAllThePies' to 0.8
>>8476220>tfw I'll have two wakfores with extra toss, plz
>>8476259>Lovecraftian
Does anybody have JoyboyAI paid content
What are the best places for downloading LoRAs for NSFW images?
>>8476445CivitAI
question to flux users. guys, did you manage to do sexy poses for flux without loras?
>>8476560Flux is mostly people standing around looking at the photographer in my experience. Had a lot of trouble getting any one to do more than neutral/photo poses.
>>8476450>>8476501>>8476503>>8476515Finally something worth saving
>>8469046I downloaded deepface lab. Looking at the tutorial but a bit confused. I want to take some jpeg pictures of I have faces and use them on porn videos, but the FAQ is for taking things from video to video. Any steps I can skip or need to change?
>>8476754all the images in this topic were NOT made with deep face lab.you are better off and i mean this very seriously using youtube. there are so many step by step tutorials for this kinds of stuff.
Has anyone managed to create realistic wallbutt or sex machine (the bulky hentai type) images? I am too embarassed to post anything, but I have comical amounts of trouble with eliminating gieger type interfaces between the body of the character and the objects she's supposed to be tied to.
>>8476886Seems to be a lot of programs to choose from. Mostly want to take porn vids and put other faces on them from images I have. What would be a decent one to use?
>>8476964You probably have to create a lora for this from photoshopped images.
>>8474527>Ask yourself this: It is X times slower than SDXL, but do you need more than X attempts at a perfect gen?But it's A times easier to prompt, B times easier to train, and C times higher quality outputs.
When Pony 7 drops it might surpass Flux, but v6 is still great.
>>8476594Made with FLUX.Schnell online, no LoRAs.>tfw the English language is your friend
I never see Missionary poses, eg;>ugly man doing press ups over french maid avec large knockerskind of things.Do they always turn into Lovecraftian Monstrosities or something?>asking for a friend
>>8477341ControlNet is a must-have add on for a reason.
>>8477352What does it do?
>>8477136you can play around with swapface program. it will have a watermark, but you can easily edit that out with adobe premier. but if the video has alot of mouth opening, it does teeth bad like many of these programs.maybe its better if you pay for it and get to use their high quality tools.ive only really seen deep face labs do face swapping good with the target is opening their mouth often which is most videos.but you need to watch youtube tutorials for deep face lab there is a lot that goes into it compared to swapface
>>8477356It allows you to use an image or wireframe as a reference for your pic.https://github.com/Mikubill/sd-webui-controlnet
Can a wizard make some nudes using her as a model? Her name is Serena Mello
>>8477378very realistic bravo. can you make Emilia Clarke schoolgirl ?
>>8477378Bravo, bravo.
Is there a way of making several positions from the same character?
>>8477399>gen first image with prompt sit>gen second image with prompt stand>gen third image with prompt from behind>gen fourth image....
>>8477408then combine that images into one file. and then img2img that new made combined image and when it gens that img2img it will be the same character most of the time with the same clothing if they arent naked
>>8477424kek
>>8477424Damn thanks bro!
What do you do to get really consistent characters like seen throughout the thread? Which trainers, how many photos, any tricks? I made my first lora using about 20 photos and the Google collab, but it only sort of resembles who I'm trying to replicate. Basically, she has the same eyebrows, glasses, and hair, but the rest is always different.
>>8477444May there was a problem in captioning. 20 should be enough to get a reasonable likeness but probably with little flexibility..
>>8477445>>8477447No futa, see OP.Go to >>>/aco/futasdg
do mods enforce made up rules?from people that post this?really?
watch out everyone this newfag >>8477452 is trying to get this thread closed like that flux thread that was up for a week
>>8477387
maybe someone who isn't a retard and loves promoting /b/ threads can make the next thread.
>>8477459this one is even better
>>8477452It goes explicitly against the thread theme, violating Global Rule 6.
>>8477479Since you love quoting rules, maybe you can quote the one about realism.
>>8477481This is a generalist thread.Go report all the /a/ threads and material while you are at it.Take your posts to the futa thread or /b/.
>>8477479>All images and discussion should pertain to 2D/3DCG>Images depicting .. generally seen to be "extreme" in nature are not to be uploaded.> All images should be of high quality and high resolution. No doodles or garbage artwork. lmao>The posting of 'real' images is discouraged.
>The posting of 'real' images is discouraged.>discouragedGuess /b/ is too slow today.
>>8477491did you get a big thesaurus for that reply?
>>8477459>>8477378damn flux is starting to get good
>>8477496>imagine not knowing any words that have over three syllables>economic collapse underway>job market demands people are educated and eloquent enough to describe a scene both concisely and preciselyNGMI
>>8477526Wait until you see the animated bestiality images it can generate.
>>8477444Training takes some massagingSounds like your learning rate is too lowIncrease it or train for more epochsAssuming there is no issue with your captions (all images should be tagged with the trigger word)
>>8477537>>8477545>herd of North American Bison in their natural habitat?
>>8477547I do not understand the difference between under-training and over-training?How can you tell what magic numbers to use?I've watched a few 'training vids' and it seems like;>I DID THIS AND IT WORKSbut they never say how they home in on those numbers.I've never tried it. Maybe it makes more sense if they're doing it?Sounds like something I'd love to learn.
>>8477553There are tell tale signs of each, refer to the guidesShort story: Undertraining shows itself by lack of likenessIf you apply the LoRA with a higher weight than 1.0 and it gets better, you are closeOvertraining shows as visual defects when UNet is overtrainedAlso it starts to reproduce the training images closelyWhen the text encoder starts getting overtrained, pieces of clothing or poses from the training data start showing up a lotWhen using a LoRA, you can set weights for UNet and TE separatelyThat's a great method for refining your learning rates
>>8477553Under training: Doesn´t look like it anythingOvertraining: Will get the *same* result on every pic.There are sweet spot ranges, but depend on a couple of factors, thus why its recommended to make a couple of saves in the LoRA´s creation process and test them.
>>8477565>>8477571Thant makes more sense now. Thank you.Often times what must seem obvious to these guys just needs a little bit of clarification.That makes FAR more sense now.>Under means you have to force it more than normal>Over means it is simply outputting the images you're training it on rather than using them to simply modify the outputThat's really helpful, guys.Thank you.
>>8477622Looks like those Aztec PCs were really heavy.
One thing that always confuses me about AI images is I can understand how it produces ONE figure, say a woman with large boobs, but if you want a guy in the image, from what I've seen and from my small experience using online tools, the guy often ends up with the same hair, big tits and her clothes too.Is that the advantage of using a better interface like say ComfyUI?Do you have one pathway for the girl and another for the guy and once for the scene THEN they're merged into one only at the end stage?Or am I way off the mark?Is that what all those extra nodes and tools are for?
>>8477459MORE OF EMILIA???
>>8477647Has nothing to do with comfyLook up regional prompterSomething like that can be achieved with comfy as wellBut it's not needed at all if you have a model that is trained on images of men and women togetherPony for example has no problem generating these scenes
>>8477668Ah... right.Thank you.
>>8477924>>8478014>body horror>face horror>fucked up hand>most basic backgroundFlux looking worse than SD 1.5 here
>>8477553Use Prodigy optimizer and it will automatically determine the best learning rate.
>>8478041>body horror>face horrorwhat, where>most basic backgroundit's an office space>fucked up handthey can't all be winners>Flux looking worse than SD 1.5 hereabsolutely not, I've gened enough 1.5 to know
>>8475204can you repost these please? the site is giving a 404, or post a longer lasting catbox please.
>>8478384First face looks derpy, second face isn't even photo realistic (plastic doll at best)Nipples are horrorFirst image skin texture just looks like camera grainSecond image has so little skin detail it looks like a drawingThere have been way better 1.5 gens posted, takes some skillXL is easier to get good output without skill, but lacks detailWhy flux gens looks even smoother than XL is a mystery
>>8476107Would you share any example prompts for a noon like me to analyse and replicate?
>>8478551>Nipples are horrorBlack Forest Labs fucked with nipples hard, by default it creates horrible red meaty circlesloras fix it but then introduce their own bias so it's a balance>Second image has so little skin detail it looks like a drawingwas messing with the lora stack, the base plasticcy look came through in that one>Why flux gens looks even smoother than XL is a mysteryeven these?>>8474666>>8473242remember you're comparing a base (+loras) model vs models that have had finetunes and several iterations of loras to focus on people and nudityhave this super powered creep
>>8477450nah, cry about it
>>8469162what checkpoint?
>>8477658flux is free for all, dude
remember when all the brown people hated ai images and now they are the ones making and posting them the most
>>8478745best part is the browns overused the word slop for anything and everything and that word reached the point of critical mass that it annoys most people and the word is dying out.
>>8478746you can always rely on a brown person to take a meme and run it into the ground
>>8477408Nice, th>>8477399Nice, thank you !
biglust is actually incredible
Still toying with things but i think im getting a hang of the work flow
i found the mini foot chair to be humorous on this one
>>8478575>clipping through cubicle wall to grope an unclothed plastic-faced robot with rubber nipsJust another office day at the Lovecraft Corporation Inc.
>>8478745>brown people can mathOh... you mean Indians.
>>8479121>you CAN buy those on Amazon
New thread for when this one falls off.>>8479443>>8479443
new to the realistic /aco/ side of things, what model do i start out with?
starting to get a hang of this ai thing
>>8479025PLEASE make more hucow gens!
>>8469166Amazing, imagine this with a better model and tools. You could create your own animated by AI film.