got flux to work with 4060 8GB lol

>>45311 i ended up deleting the pics, was fucking with the board spammer yesterday and would say this is the one thing that i did where i went a bit too far, if you had any progress on that request, would say not to post them

I think it was the original thread that the OP posted a dataset for training a LoRa, that link is dead and I wished I saved it, but if you're still here I could use it

Can you share your settings and flow? These look so awesome and fucking good

The earlier posts arent mine but look to be simply using Flux dev and the pullup lora on Civitai. I've mixed in a few other lora's but the output differences are subtle. Can do a lot if you just play around with the prompt and cant go wrong cause its flux.

>>45423 Can't go wrong with generation, but 2 images of the same damn belly button ring poisoned the whole lora.. I miss negative prompts.

Can I requesting two giant size male babies based on this image, between the ages of 1 and 4, both about five feet tall, with one baby having short black hair and black eyes and one baby with short brown hair and brown eyes, both with slightly chubby bodies, and wearing only plain white diapers, standing, full body, smiling, anime style. Keep it SFW please,

Too bad you couldn't make it to the convention this weekend anon.

Requesting an anime style couple, with one male age 30 with short brown hair, brown eyes, a slightly chubby body, and wearing glasses, and one female, age 28 with medium brown hair, brown eyes, and an average female body, both shirtless, wearing plain white adult diapers, wet diapers, standing, and smiling, both embracing and hugging with the male having a visible bulging erection underneath his diaper.

Marvel Rivals has awoken something in me.

Made some more art myself.

When mommy and daddy decide to become babies again...

>>46256 Honestly that looks like fun. I Wish that would happen to me.

>>46251 sublime

>>46256 thats awesome anon, could you share which prompts you used? would be lovely to see a video of this

indian diapered inpaint

>>45499 Me on the left

new illustrious plain white diaper lora on civitai for u all :)

>>46755 The headpatting one really does it for me

Inspired By Tykables got an absolutely frosty reception so I wasn't really planning on following up, but my studio got flooded and is still under reconstruction and I got bored. There's now an Inspired by ABUniverse set of loras. The LittleKings and BunnyHops are nothing compared to what Miku has done, but the datasets were already made. In other news, Flux sticky tape diapers are still causing me problems with the standard set of photos I took. I think I need additional images compared to pullups and hook and loop diapers as adjusting repeats and hyperparameters has not improved the consistency. More to come on that once the contractors finish painting and stuff. I also acquired an additional "model" that can fit into a Pampers Easy Ups. Looking forward to adding the My Little Pony ones to the Pullup lora set once I can take photos again. I'll get a few samples of size 8 diapers to see if they fit. No idea how I'll be able to get something posted on Civit. They flagged the yellow shirted anime girl and I can't image how I can get proper baby diapers on a realistic woman past the SJWs. (That's the reason the anime girl has enormous titties now.)

>>46755 Illustrious with or without your lora is still pretty good. It really sticks to prompt adherence.

>>46997 what language/prompting are you using to get quality messing gens like that?

Anyone know where to get the japanese diaper/pullups lora? Like merries, mamypoko, etc like the one used by this AI art account

It's crazy how far this stuff has come in such a short time frame. Even crazier to think how much more polished it's going to become in the years ahead.

>>47036 My prompt for the first one was "1girl, clenched teeth, day, bathroom, yellow diapertapes , blush, squatting, clenched hands, eyes closed eyes, skirt, black hair, blurry background, depth of field, miniskirt, standing, bathroom, pantyshot, short hair, twintails, curly hair, looking at viewer, blurry, messydiaper, action shot, headpat, stink lines from butt <lora:Plain_Diapers_Illustrious:1> Negative prompt: worst aesthetic,worst quality,old,early,low quality,bad quality,lowres, scan artifacts,jpeg artifacts,signature,abstract,lossy-lossless,username,logo,watermark,abstract,ai-generated,greyscale,monochrome,sketch,unfinished, (bad hands:1.2),mutated hands,bad face,bad anatomy,(missing fingers:1.2),extra digit,fewer digits,censored, (simple background:1.3),overweight,fat,skinny,ugly,(chibi:1.1)," There's an auto prompt plugin for forge that helps a ton if you are using a model trained on danbooru tags that you should use though. Makes it easy.

>>46755 I'm impressed how flawlessly this LoRA works. The shape of the tapes is almost perfect.

Any guides on there how to get started with this stuff?

this dude is nailing it https://www.tumblr.com/foofyfoo

>>47077 I need more of this. Feels like only a year ago people were making crappy Dall-E images as a joke. At this pace even a year from now it's going to be insane.

>>47109 https://youtu.be/g74Cq9Ip2ik If you want anime style images, look for illustrious-XL checkpoints - if you want realistic, then Flux is the better choice.

>>47075 https://runya062.booth.pm/ https://www.pixiv.net/en/users/19754579

>>47232 Thanks a lot man! sadly they mentioned does not work well with pony. only Illustrious

Trying to experiment more with the Kling image to video generation. Not exactly what I wanted, but still worth sharing.

>>47265 Silly me, that’s the image. Here’s the video

>>46994 Target Up and Ups size 8 fit! Unfortunately I think it's going to take weeks to rebuild the studio before I can get more photos. While I was looking for samples on Ebay, I saw some Japanese samples. >>47075 I'm not at all familiar with brands. Can anyone recommend ones that would appeal as a lora and would fit? My 'models' can manage a pampers easy up and above. It'll be Flux first since my Illustrious loras are meh.

>>47277 Probably what i mentioned like merries, mamypoko, goo.n. They’re big in JP and have lots of pics on twt. Trying to train one myself and see how it goes

>>47306 Holy based turbo autismo. How you found someone willing to do photoshoots is beyond me. I kneel.

>>47323 That's a piece of plastic

>>47252 In my opinion Illustrious is far better than pony. There is no need to stay with pony unless you are doing some furry stuff, but even tho I'm pretty sure there are some Illustrious checkpoints that do it right

>>45868 I did it myself.

>>47355 A few more.

>>47329 The eyes are censored

>>47350 Oh wow! The videos seem to work really well! That's fascinating!

>>46755 I'm new to slop generation so forgive me if this doesn't look great, but I'm pretty impressed with the lora so far.

Miku chilling in lemon soaked jucie diaper

Target Up & Ups worked in my makeshift temporary studio. Just need to pick a epoch and run some publish images.

Anila is a qt

>>47516 Are there any baby diaper LoRas published anywhere?

>>47800 'Diaper Styles - Tiny Diapers' on civitai for flux.

Hopefully I'll get the photos done for the merries that are currently on the model this weekend. Unfortunately, baby diaper patterns don't have names. Anyone got a better name than "Merries Neighborhood?"

what kind of model and prompt are you people using?

>>47930 BRO IM SO EXCITED FOR THIS

sometimes it comes out neat sometimes it makes you want to claw your eyes out

>>48125 Holo the Ditzy Pup

any more bedwetting ai art ? feels like we barely get any sleeping content.

I made some chubby gay ABDL art.

>>48195 .....WTF is this shit? Please, for the love of God, don't bring up this faggot shit again. Seriously, who the hell finds this shit sexy?!

>>48201 >who the hell finds this shit sexy Gay guys I asssume

>>48201 It's garbage, but the fact it's like 40 fucking images of the same thing is the real problem.

>>48201 I'm not attracted to men, but I'm not angry about other people being attracted to men. This is weird but given the nature of our fetish it's bound to happen.

>>48220 thats literally everything with ai though

>>45317 > Illustrious pleaseeeee share what main checkpoint model and lora checkpoint and prompts you are using for these!!! <3

>>48290 i'll tolerate dudes in diapers if they're somewhat attractive. these are all skinnyfat nerd virgins. actually, a hard cock bulge in a diaper is hot, but just make the guys have abs and be in shape and not look so fucking bad.

Am I able to create something with my current setup? 1660 Super I5 11400F 16GB Ram If not, is there some reliable cloud thing like the one from google? I saw someone using it a long time ago but couldn't get how to use it properly, it would be ideal if I could run it for like a hour or two just to create some images, I've been using NovelAI to make some chat bot interactions and I wanted to create some images based on those interactions

>>48372 Probably, just grab webforge UI for flux.dev/illustrious or foocus for anything older than that. They are both one click installers. Your card is old though so it will probably take a few minutes to generate anything on any newer models.

aight whos gonna try the new openai 4o image gen

>>48390 That one part in the Snow White remake where she has been cleaning for hours and her diaper is too full to take anymore from her, then the evil queen arrives to tell her its time for a change and that step-mom knows best, to which Snow White reluctantly agrees, and then is happy that it happened.

>>48290 Not OP, but it's flux and the pullup lora on civit. Prompt is likely just a description of what is shown. "a girl on a bed wearing a pullup" or something like that. >>48399 It would reject the request. It's not open source or local.

>>48390 What would you recommend? The ones you posted look pretty good although i'd prefer a more anime look, is the difference between webforge and foocus that big? And do I risk my GPU using it too much to generate stuff? I'm comfortable with like 2 to 3 minutes of wait, anything way longer than that would start bothering me Also worth mentioning I would need to use my HDD since my SSD is almost full (bg3 is a bitch)

>>48493 Foocus is easier to use but it can't run the flux or illustrious models so you will be stuck with sdxl or lower models and no one is really making loras for them anymore. You really need to just start using it to figure out what you want. civitai has a ton of diaper loras you can use. TLDR: Just fucking try one of them.

>>48499 Tried ComfyUI since I saw people saying it was the less heavy one, pretty nice but really it takes too long and the UI is a bit weird to deal with, also tried webforge UI with flux.dev but didn't try diapers with it (just a simple cat photo took 15 minutes), ComfyUI takes around 13 to 15 minutes to make one image for me, I used Diaper Styles from Civitai, I'd like to know what LoRa you used for the Pokemon and the Alice ones you did, I liked their style especially the Alice one

>>48509 15 minutes is ridiculous. There are ways of getting flux to run well on very low vram gpus. Google flux and your gpu to find instructions (or ask ChatGPT). Words like Quant, FP8, GGUF means you are on the right track. All the anime ones are going to be illustrious based models which is itself an SDXL based model. It should run much better compared to flux which is an enormous model in comparison. NoobAI and WAI-NSFW are great illustrious fine tunes that I would recommend trying first. They can do diapers and characters normally, but there are plenty of loras available for more control. If you are getting cloth backgrounds from the loras, the weight is too high or the description is too short for flux. Flux likes very wordy prompts.

>>48509 Any good anime AI you are seeing is probably done with illustrious which is just a heavily trained flux. It uses booru tags and there are a few diaper loras for it on civit. With my 8gb 3070 I can do illustrious or flux generation in a minute or two on webforge and sdxl generation in about 30 seconds or less on foocus or webforge. Assuming you don't have a potato PC do some searching for the best GPU driver for image generation. There used to be an old driver you needed to use.

>>48531 Why do you call it flux? Pony, Illustrious and NoobAI are all SDXL trained models.

>>48532 Oops, yeah that is an sdxl model my bad. In any case it's better than pony models IMHO although both can do not bad diapers without a lora.

>>48531 Kind of managed to do it, the faces look horrendous though, had tons of problems with flux over something to do with clips and VAEs, I tried using Afrodite for that but yeah no chance even had a blue screen once These are from waiNSFWIllustrious_v120, it took me a while to figure out Pony prompts my friends used to use aren't the same as Illustrious ones, it took me 3 min to do these the prompts were: masterpiece, best quality, amazing quality, very aesthetic, 1_girl, diaper, teenager, auburn_hair, camping_tent, full_body, small_breasts <lora:Plain_Diapers_Illustrious:1> Negative Prompts:worst quality, comic, multiple views, bad quality, low quality, lowres, displeasing, very displeasing, bad anatomy, bad hands, scan artifacts, monochrome, greyscale, twitter username, jpeg artifacts, 2koma, 4koma, guro, extra digits, fewer digits, jaggy lines, unclear Any chance I can improve on these? The faces especially the eyes still bother me quite a bit and I can't seem to figure out how to do different styles, there were some really good ones on the White Diapers Illustrious homepage on civitai

>>48547 Not exactly sure with WaiNSFW, but the base Illustrious model is trained without using "_" for tags, so adding _ reduces the efficiency of the tag and can even remove it entirely. For styles you can prompt artists directly and you can of course prompt for stuff like watercolour and photorealistic. Different models can also have inherent biases towards certain styles if not prompted otherwise. Also unfortunately anything below 768x768 will have reduced quality as the models have a base training of image resolutions 768 and above.I like to use 832x896 for example. (also I prefer DDPM as sampler and "Align Your Steps GITS" as scheduler, but most choices do well and it's really just RNG when one works really well with a certain prompt and seed and when it doesn't. Just avoid "DPM++ SDE", it has the best quality but the worst speed by far) In many cases not having stuff in the negatives can be better than having a lot, or simply just using 1-3 things. Example negatives to test: 1. [black:white] dress, clothes pile, close up 2. gym, white dress, tucked in shirt, frilled apron, bloomers, bow panties, suspenders, harness, inflatable toy, print, stickers 3. bedroom, bed, 3d model, hanging panties, diaper mattress, napkin, body pillow, pillow, lunchbox, (frame:1.4), toilet, square, rectangle 3,5. bedroom, 3d model, hanging panties I wouldn't use "teenager" either, at best "teen" but if it isn't specifically loli/shota or mature/milf then just describe the body attributes you're looking for without mentioning age-range as it's not trained on those. You can then put unwanted body/age results in the negatives. Boobs too big? Put big breasts in negatives. Body too young? Put loli/toddler/shortstack in negatives, etc etc

>>48547 Here you can see some examples of how volatile AI can be and how there is never really a definitive answer to anything. In the first image you can see how wildly it can differ when changing up the prompt in relatively minor ways. In the second you can see how much the Lora itself can influence and "overtake" the image in quality, style and composition. Thirdly you can see how it can change with different negatives and a resolution change. And what works well for one model might not work as well for another model despite having the same base.

>>48552 Actually pretty good now all changes imrpoved it a lot, still not able to do a bunch of styles though just these more generic ones and the faces and diaper prints are kinda weird if I look up close but definitely better The changes increased the time a bit though, I was doing the images in 3 minutes before and now they take around 5 to 7 minutes with increased resolution and different sampler and scheduler I think the real problem was the Plain Diapers Lora though, it was messing up the art even after the changes at 1 strength, first two pictures are from real-ish diaper 2.0 with 1 strength and 3rd and 4th are from Plain Diapers but with 0.5 strength and some of the negatives you recommended Still not able to do stuff like the alice one from above post nor the one from plain diapers homepage though ( 5th and last image)

>>48562 Also, managed to get flux to "work" now instead of blue screen or error message I get 40 minute generations for a blank or completely blurry image So yeah, don't know how to make that one work yet

>>48562 Yes naturally it's a big increase in resolution so it will take longer unfortunately. What interface are you using? Forge (and ReForge) are based around being optimized and fast, but there is no fixing slow hardware in the end. https://github.com/lllyasviel/stable-diffusion-webui-forge If you mean Alice from >>48499 that's a lot of PieceOfSoap's style, whilst the facial features has more of a... I can't remember the name but I''ve seen that smirky mouth and pointy noise combo somewhere Piecofsoap is not really trained on in Illustrious, but the plain diapers lora is clearly trained on pieceofsoap, so you would just have to experiment around or hope that guy tells you. There are some illustrious models with a more western bias to their default style too.

>>48562 To get that Alice pic I used the plain diapers lora plus an art style in my prompt to modify it. Maybe disney? I forget since I just do a ton of batches with a ton of different styles. Illustrious is trained on so much drawn art you can get wildly different results just by changing artists in the prompt. There are plugins for webforge that help with tags as well. The one I use is called webui-tagcomplete I think. https://mzmaxam.github.io/Illustruous-Artists/index.html shows a bunch of different artists with the same prompt.

I did some art of an anime diapered father and son, but I don't know if I should post it on here, because it has shota.

>>48778 i really which there were more vintage clothes in ABDL; garish 90s colors with barney, going all the back to 20s gowns with big puffy and sleeves and peter pan collars

Photo studio is still under repair, but since I've got a week or two before any more work is done I've moved some stuff down there to take a few photos. Training a Parent's Choice fox diaper now. I also published a few Illustrious loras for LittleForBig since I had those photos already. With any luck, I'll be back at full capacity in a couple of weeks.

>>49242 God bless you, anon. Whoever you are, I thank you for your service. I hope you are successful in life.

>>49316 Thanks. I do it all for likes and emojis. HiDream is a new model. Comparable to Flux as far as size goes. Not great at diapers right out of the box, but clearly better than Flux. MIT license so it should get some love from the community. Here's hoping it's as easy to train a lora for as flux is.

>>49316 Seconding this.

I made an Abena M4 diaper lora on Civit, first time making a Lora. It handles wetness fine, but it doesn't always trigger. I'd love to make a model that handles masturbation (hand inside front of diaper) next, as well as tackle other brands like Cushies, Tena, Bambino, etc. https://civitai.com/models/1481794?model

>>49462 That one turned out really well. Good job! Unfortunately your upload coincided with civit absolutely crapping itself, so you are going to have to shoulder some blame for that.

Civitai just banned diaper content. download my models now before they get deleted in 30 days

>>49577 Disappointing, but not unexpected. Their filter blocks the word baby which tripped me up training on their site. Since I won't be able to use their site to train loras does anyone have a good guide to training on 8gbs of vram? Hopefully another site pops up that does allow abdl content.

>>49577 >>49462 Well shit, I just made a Tena one as a follow up to the Abena. Get it while you can: https://civitai.com/models/1500041 I will more than likely continue on https://tensor.art but I'll need to check it out.

>>49588 8gb is garbage tier for ai to be frank. You might be able to train locally with block swapping. Kohya_ss using Bmaltais's gui is my preference. End of the day, anything less than 24gb of vram is going to be subpar. Renting a gpu is probably the best option.

>>49604 tensor.art is not exactly a solution: 3.6 We highly value the user experience as well as the safety of our community. Therefore, we strongly discourage users from posting, sharing, or spreading any content considered NSFW (Not Safe for Work). This includes, but is not limited to, pornographic, violent, insulting, or other potentially offensive content.

>>49608 Well damn, I only said that because I saw it suggested in comments. Where should we go? Regardless, I'll be sure to always post future loras here.

I don't suppose anyone's going to make a real wanx diaper lora now are they

Someone else suggested going to https://tungsten.run/ Uploaded a Cushies Lora to Civit still, get it while you can. I'll make sure to put these somewhere else when I can. They are already removing my generations from there. https://civitai.com/models/1509763?modelVersionId=1707808

>>49685 There's a discord link going around to find a new place. You should get it soonish.

>>49685 https://www.diffusionarc.com/ was posted to reddit. Assuming it has similar features of civit it should be a suitable replacement.

Realistically an easy way for now to archive this stuff is to put it on huggingface, like this guy did a while back; https://huggingface.co/toasterroasterp1/ABDL_Pack/tree/main these are outdated but someone else should do the same thing. huggingface provides completely free hosting of any amount of stuff. then you can even use this site to search for it all: https://c*v*t**archive.com/ (replace with vowels, its the name of the current site everyone uses.) i'm not going to do it because i have a strict "no using my email account for jacking off" rule. but someone should sign up and upload everything. ideal would be to throw in some information about how popular the model was into the filename so people won't have to sift through the 75 weird loras of morbidly obese photorealistic women in diapers.

>>49685 Did civit nuke this or did you end up deleting it?

>>49796 >>49462 >>49604 >>49685 They nuked all three of my models and suspended my account. I think I was the first to go because there are still diaper models on there but slowly they appear to be dropping. Like >>49703 suggested, I put them on Huggingface and if I put them anywhere else I'll chime in: https://huggingface.co/Adamtwo22 >>49686 I got the link on Civit thanks, just haven't had time to get in and check it out.

>>49811 They unpublished a couple of my models. I think any one that had a messy diaper post under it got nuked.

>>49811 >that last pic >we might live to see "Nappy in the Shell" animation come to life

>>49816 Well, they didn't ban these, the OG messy diaper loras: https://civitai.com/models/88065/omorashi-omutsu-yuzu https://civitai.com/models/95360?modelVersionId=306604 >>49851 Yes, I love her :3, but the Civit police arrested her :(

>>49871 What a shame

All of my diaper LoRAs were deleted from civitai, but you can still download them from huggingface: https://huggingface.co/MikuMasterRace

Anyone know if we will still be able to train diaper loras on civit? Their lora trainer was cheap and really easy to use. I don't really want to fuck around with other sites or try grabbing a GPU if I don't have to..

>>49923 as long as it doesn't need to use the word diaper

>>49924 I guess I should have tried, but yeah it flags every caption with the word diaper in it Hopefully another site with an easy to use lora trainer pops up to replace civit..

first real crack at a video lora, not a complete disaster shame I have nowhere to post it now

>>50005 Throw it on huggingface, that seems to be where a few people are putting their loras.

Can you guys make lolis in diapers?

>>50005 Nice, one of the few fapworthy IA slops.

>Get back into prompting <Civit has band diaper loras Fugg. Regardless though, what's the current "prompting meta"? Looks like Illustrious is better than pony for anime, and flux is for real images. How good are these models at characters out of the box? >>45497 and related make me want to gen some diaper cosplays. Should I be redownloading a bunch of character LoRAs trained on Flux? Also, what's the state of video genning? Last I looked into it some sites had models up, but local videos seemed like a bit of a pain to set up (on linux at least).

>>50198 For Illustrious just use booru tags, and for flux just use nice descriptive sentences. Illustrious is great with characters out of the box, not so sure about flux although loras exist and you will need them for diapers. As for video just use wan 2.1. You can run it locally via pinokio or online if you are ok with watermarks. Runs decent but needs power to do more than 5 seconds worth of video.

Did some stuff of Nasa Huston from Orbital Children.

>>50210 WOW! These ones are freaking amazing! Could you please make 5 pictures like these of Mina Misasa too? And then another 5 of pictures like these ones of Konoha Nanase?

>>50205 Thanks Anon, now I too, can AI slop. Though, I suppose I'll have to set up comfy UI. Is it as much of a pain as it looks like? Also, eyes are hard if anyone has any suggestions for getting them better.

>>50198 >Civit has band diaper theoretically could you circumvent banned words with something like 'bulky disposable incontinence underwear'

>>50217 I'm not using Civit to generate, just to download LoRAs. They were more convenient to get LoRAs from and it was nice to have a centralized location, but c'est la vie I suppose

>>50218 im talking more just like in general

I just saw some pics where they're wearing something like shorts over the diaper loras? how is that possible?

>>50214 I cannot make any AI stuff of any other character from that series. She's the only one available.

Hey guys, I dived into image generation two years ago and I had a lot of fun with creating loras with python scripts and messing with stable diffusion when it was brand new. However, this was time consuming and gave mediocre results 95% of the time so I decided to shelve it until it got better. Can I have a tl;dr of what changed in two years? What's the go-to solution to do local image generation and loras training now?

>>50337 a1111 used to be the most common inference tool, now it's 'stable diffusion webui forge'. i find forge-attention-couple to be the most helpful extension. noobai, and to a lesser degree illustrious, are really good if you just use a lot of booru tags.

Given that it's basically impossible to browse loras now, but a ton are on c*v*taiarchive, we can talk about which loras are there, and what they are good for. i will pick a couple that i like a lot https://c*v*taiarchive.com/sha256/3d53dd949137124f9df70e4878b0adf44012d43f6252a36890917d152de474c6 diaper change lora, this is very good, can set it to 0.5 strength and (diaper change:1.5) or so and it will work really well without changing style much. https://civitaiarchive.com/sha256/43c3f21bcd60b1205ec08967933a81c523e8e9be6eea04c3d63c1fcaec0e59f0 does the same thing, is used the same way, and might be better. https://c*v*taiarchive.com/sha256/06bab78c2f9b7a6349f48e6ddb5cc8d3492e2eed9cdeeef5c1bca69453436368 although this is just labelled as "diaper" it actually seems to be designed for illustrations of diapers pulled halfway down legs. https://c*v*taiarchive.com/sha256/27c681b0a607751f65b684dd2c7111b6ba438f1e4351eda4e7dd6d80c18eb2a7 for a really thick diaper with no styling. creates a sort of vase-like shape

>>50345 wtf, that's ai?

>>50337 >>50341 Idk anons, I've been getting pretty comfypilled and I'm starting to like it

>>50344 Whatever you're doing do more. I really like the maid one.

>>50349 What's this software?

>>50362 ComfyUI. It's a visual programming interface for general AI generation. It's a little overwhelming at first, but it gives you so much control over the process that with a little bit of tinkering and tutorials I'm starting to like it.

Started getting into img2video - my second attempt didn't come out "too bad(tm)"

lowkey addicted to putting hot celebs in diapers 🥵

Somebody please make diapered lolis with ai, you're my only hope left.

>>50422 just make an account on pixiv, or if you mean photorealistic, kill yourself

civitai might just die within a year or two now. there are torrent sites in case huggingface starts deleting these: https://civitasbay.org/ their search goes straight to civitaiarchive.com looks pretty great!

>>50437 Nothing realistic, just cartoon lolis, sorry for any confusion.

>>50422 Here's a batch. It's not hard to make your own though...

>>50456 Great, thank you.

>>50344 These are really good! Post workflow please

How can i get the nipples right in flux?

>>50472 Use a better base model. The Acorn is spinning flux model on civit is trained on just nipples. Not sure how it performs with loras though.

The results from Veo are in. It's not free, and the best model (that also can generate audio) is pretty expensive and only generates audio half the time. Fed some of my generations into it from the loras I made. It's successful about 70% of the time on Veo 2, and only about 30% of the time on Veo 3. Some images I tried just continued to fail no matter what, and other images eventually went through after retrying it with the same prompt a few times. Anything remotely explicit will likely be shot down immediately, and images of real people (famous enough) will be immediately rejected. Also, you can technically download an "upscaled" 1080p version of your creations, but every time I tried, it said it failed due to a policy violation, so I am stuck with the 720p versions. I'm looking to find the tricks to improve successful generation rates. Best of luck if you try it.

>>50481 Here's some more. The audio did not generate on some of these.

>>50481 That DJ one is really good for AI. Not too many cronenbergs in it. I'm sure china will release a more open source one of these that we can run on consumer stuff.

Is it just me or did Civitai purge all ABDL/Diaper content?

>>50492 yep, see: >>50343 >>50198 they say its because credit card processors asked them to get rid of the "gross" porn stuff. they took down various things like diaper, vomit, pee, and they just took down celebrity loras. then the credit card companies said they wouldn't work with them anyway (think tumblr). however they are keeping all that old "gross" content banned, which suggests that they wanted to do it anyway. i guess i can't blame them, furries and futa is still on there and they are relatively common fetishes, but diapers still freak a lot of people out.

I was able to save a fair bit of diaper loras before the purge. I've been fiddling with comfyui for a bit now but I don't think I'm especially great at it, and my PC isn't the greatest either. But, if people are interested I'll take requests or post more when I have spare time. I haven't quite got the hang of multiple character images so if anyone has tips on how to get specific outfits/hairstyles/looks and not generate two characters that look identical that would be awesome.

>>50481 >>50488 As they say, this is the worst it'll ever be. Look at the first AI thread versus this one as far as improvement, then imagine one year from now. I've already replaced probably 70 - 80% of my diaper content consumption with AI diaper content instead of real women. It meets my tastes, and I can always just generate a new idea whenever I want of any subject I want, kind of crazy. This being open source will be a nuke to adult content creators (already making an impact), and I can see how AI is going to wipe us out in the future. Enjoying the ride though.

>>50498 Multiple characters can be hit or miss. I've had some come out decent, and others need a lot of inpainting to work with. this one, for instance, I needed to redo the hair style on the left girl, and remove glasses from the girl on the right. My best advice is to prompt like: {Character 1, Descriptions, and, Tags}, {Character 2, Descriptions, and, Tags}, Character 1 AND Character 2, The, Other, Tags, You, Want, to, Prompt For some reason, I've found it works "decently"

>>50519 I too enjoy this era of Free Porn

>>50592 >webp did you mean for these to be animated anon? You should probably convert them to webm

>>50578 Using the Invoke UI running locally I've found using regional prompting generates reasonable results for multiple character. Including things like "3girls, multiple characters" in the base prompt and then describing each character in the regional positive prompt.

>>50594 They aren't, you can thank using Pixai for them being webp... I think.

>>50595 >That image Very hot anon! I haven't tried regional prompting on ComfyUI yet, since it basically didn't work on Automatic1111, so I'll give it a try there too. Did you use an Illustrious model for this image?

>>50627 Yeah using WAI-NSFW-illustrious-SDXL v11 as the base and then both "[ILLUSTRIOUS] Car seat" and "Safety Mittens & Restraining Booties" as LoRAs. I find with Illustrious try and find a few artists you like the style of and use their name in the prompt, that way you're not just hoping you get the style you like.

>>50642 Yeah, I've been trying to recreate your picture as practice, just using different style LoRAs I guess. Or are you not using Style LoRAs and just using artist tags?

Having a lot of fun with Veo.

>>50658 Fucking dystopian. Very clearly AI right now, but I can imagine some Jew using this tech to cut corners. I'm currently a college student studying acting, and I hope to Christ people recognize the value of art being a humanistic thing that no other entity can emulate.

>>50659 >Fucking dystopian On the contrary, it is very much *utopian*. I'll take slop over a world where some theater kid that fails their way into a liberal arts degree can browbeat others with any sense of self importance (such as yourself, if that wasn't obvious).

>>50658 That first vid was amazing. I'd pay money for photorealistic diaper commercials. This stuff is still just a little too weird for me to fully get into. The diapers don't look good or there's just too much AI strangeness.... but goddamn it is so close. Like if it's here and if this doesn't get banned or completely destroy human society in a few years it'll be a golden age of content.

It took me one million hours on tensor but I finally made something semi decent. Do you like it anon?

>>50658 Looks like fun anon. Before I shell out for it, got any prompting tips? How well does it handle anime art styles? How about how to get expected output? Just be super detailed like flux?

>>50666 that looks great anon

>>50662 I'll bite. I just want to pursue my dream job like anyone else, man. I wouldn't be in this thread if I didn't like AI porn myself.

Sorry for spamming, but here's an art of what an adult male from Rugrats would look like wearing a diaper.

>>50670 You are correct, just be detailed like flux. The first video was generated from this prompt: >"A young adult white woman is sitting in an adult high chair in an apartment living room, she is wearing a bib and wearing a thick white disposable adult diaper with no pants, she and is pouty and upset crossing her arms, she also has a pacifier in her mouth, the camera starts zoomed on her face before zooming out to show her full adult body and diaper and that there is a young Asian woman patting her on the head and gloating, there is also a robot maid holding a pack of diapers and subtly moving its head to interact and standing and ready to babysit the woman in the crib, the Asian woman is dressed to go out, the Asian the Asian woman says, "Dont worry, ill take good care of your boyfriend.", the Asian woman is standing next to a young man also dressed to go out, after the man hears this, he says, "Love you Angela", after he says this the two then both walk towards the door to leave and walk out the door while the woman in the high chair stays behind and looks down in defeat" Some other tips. > Describe your diaper with "no pants" added. I use "wearing a thick white disposable adult diaper with no pants" in all of my prompts because sometimes it puts it over pants. >Dialogue is not perfect and has a tendency to have characters speak based on the order the appeared in your scene, just be more detailed in your prompt or give the first character something to say to help prevent this >Anything remotely sexual will be restricted. You need to be clever. The Japanese student wetting her self in the video above was "the camera pans down to show a small puddle of water under the student's desk". You can just say "she peed herself and leaked onto the floor" that won't work Anime or really any animation is not great. If you try it, I'd recommend using the Frames to Video feature. See: >>50519

>>50695 The one with the white girl in the crib is perfection

>>50695 How often do you get "Couldn't generate video. Try again later."? I assume that's not a "we won't gen this" and more of a "we have too much traffic" error?

>>50698 Most of the time it actually is "We won't gen this." Change your prompt or just try again but there could be mulriple reasons. If anything gets past 10 - 20%, that means youve got a chance if you just keep retrying it should work. If it consistently fails under 20% that prompt is dead and try something else. Also, keep the tab active. Ive found that minimizing the window, or if youre on a phone, moving the app to the background and doing something else while its generating, has a high likelihood of failure. L Generating 2 at once doubles your chances of one being successful, but its 200 credits so... Lastly, sometimes you need to refresh the page, some gens actually complete, but dont show until you refresh the page. Its clunky. Good luck and share your results ;) >>50697 I actually wanted another scene where shes in her crib watching them having sex (but its offscreen) and shes getting turned on and starts touching herself, but thats too much for Veo :( i attached the failed attempts

>>50699 >I actually wanted another scene where shes in her crib watching them having sex (but its offscreen) and shes getting turned on and starts touching herself, but thats too much for Veo :( i attached the failed attempts We need one where she is tied up while the couple is having sex and she can't touch herself >:D

>>50699 Thanks anon, got a decent chunk of little ideas generated - though some of them came with weird belly high diapers lol >>50703 >IRL lolis Neck yourself

>>50711 These are awesome, thanks for sharing these ideas. I usually add "with their midriff exposed" to the prompt after describing their top / shirt and it keeps them from being so high-waisted. >>50709 I gave it a try in the 2nd clip, had to manually add the audio. >>50703 Fuck off

>>50658 moar

>>50731 (Like maybe a hacked version of Veo that can actually generate stuff like this?...)

I've been fiddling with tensor.art lately and I've been trying to make an image of a woman breastfeeding another adult woman. The problem is I don't know how to prompt the thing so I can describe both the mommy and the baby, and I end up having a mess as a result with description of the baby being applied to the mommy or the other way around. How do you handle two characters in an image without breaking it?

Anyone able to post the file for "plain white diapers" for illustrious? Seems to be dead in the archive

Also how do I prompt a second (or multiple people) into the pic so it doesn't get the separate descriptors all mixed up?

>>50746 Are you the anon from /aco/ that just started learning comfyui?

>>50749 Perhaps

>>50750 Check out >>50595 and use a regional prompting node. Pic related is a work flow example. The basic idea is that regional prompting allows to specify a positive prompt for a certain portion of the image. So you set up your first prompt as "a prompt that applies to the whole image" and put all your fluff (masterwork, 8k, super high res, etc.), art style, and situation tags in there. Then, you configure the region that other prompts are going to apply to, and apply your character prompts to those regions. Note, that regions don't like to interact, so keep that in mind. I'm sure there's a way you could overcome that without inpainting using controlnet - but I haven't played around with it enough yet.

>>50745 Here you go https://mega.nz/file/GswVxA6R#hDBXcuVRc2T84uRD6WOjJZ3RgYJDwywHocDhU-_NVPc

>>50675 >I just want to pursue my dream job like anyone else, man Your dream job is to become a substitute teacher? Because that's what's going to happen, lol. You will have to come to terms with the fact that the path you're on was not chosen because you are some kind of dreamer, but because you are a decidedly mediocre person.

>>50745 >>50746 >>50750 Good shit. Already on par with some of the best diaper/straitjacket AI art I've seen. Very convincing looking bondage. Could you do one in a public setting (sidewalk, classroom, etc)?

>>50746 this is a really common issue. for prompting multiple subjects, https://github.com/Haoming02/sd-forge-couple this extension can be added to a1111 and it allows for mask painting for detailed multiple-subject composition. it's a LOT easier than learning comfyui.

>>50763 I wanted to, but getting the crowd to do what I want seems pretty hard now

Ok, I'm cooking (as the kids say), but I had to redo my workflow from scratch because old one was throwing out tensor errors. Should hopefully bake some two people scenes soon.

>>50778 >the crowd For me it's just enough to have a background and that implying the scenario itself.

Is there a separate archive that keeps even the nuked loras/models? The straitjacket one isn't even on that guy's huggingface.

>>50790 Prompt?

Thank you cybersatan, very cool. >>50791 +1: 2 girls, muted lighting, masterpiece, best quality, amazing quality, vibrant, delicate linework, anime, 2 girls, loli, bouncer over diaper, hanging from ceiling, pink bouncer, black t-shirt, barefoot, facing the viewer, feet off the floor, headpat, playroom, embarrassed expression, kicking, simple irises, brown eyes +2: 2 girls, muted lighting, masterpiece, best quality, amazing quality, vibrant, delicate linework, anime, 2 girls, teen, blue t-shirt, dolphin_shorts, responsible, teasing, small breasts

From the mind of an autistic schizo lolicripplefetishist: Cripple loli physical rehab therapy (I don't know what I am talking about)

Additionaly: how do I prevent random background characters bleeding through into the image despite multiple negative prompting for no more than two characters and such?

>>50763 Unironically fuck crowds

>>50808 How are you setting up your prompting regions to get multiple characters interacting like this? Best I can get is characters standing next to each other

>>50812 There is definitely a better way to do this tho, I assume.

>>50813 Oh, you're not even using regional prompting then?

>>50817 It technically is, but only two halves and you just move the border in the Comfy Couple node. Drawing the masks might be better desu.

>>50808 Unf, these are really doing it for me, especially the first one. Thanks a lot!

Do I need a fucking lora If I want to do a diaper that is covered partially by a shirt? Anytime I use the diaper-specific loras it refuses to cover it.

>>50854 Perhaps the imagegen would be less strict about not covering the diaper if the clothing covering it was translucent?

>>50856 Ok, but I don't want to generate exclusively bedroom lingerie nightgown pics

>>50789 It's on tensor.art

Something broke in the prompt and created this masterpiece kek. I call it "social anxiety"

Girls not feeling well

Is anyone able to get AI to diaper this whore?