What model/lora/etc. is this? it looks great!

>>24692 Are these even ai? They look flawless

Thanks guys, I used EpicRealism checkpoint model and I made these loras myself training certain models like Aliss Bonython and few others and mixed them in one prompt.

>>24705 any chance you'd be willing to share those LoRAs? I'm always looking to add stuff to my toolkit as well. Regardless, that's some excellent work! Photorealism is an underappreciated side of all this, so I'm always glad to see other people playing in the space.

>>24709 Thanks! I agree here are the Loras I used for this pictures https://mab.to/t/NWyJbfTlETm/eu1 Mostly on 0.3-0.5 and you can combine multiple together, use ADetailer to fix faces and hands ;) Share your results please

>>24712 Thanks! are there any trigger words or useful prompt elements?

>>24709 God that third one is good, the way the jeans are wrapped so tightly on her gargantuan thighs is just *chefs kiss

>>24713 Wait I'll share my prompts: Positive: raw ((medium wide angle)) photograph of pakistani female ((solo))<lora:alissv1-06:0.5> with thick thighs, <lora:tayv1-08:0.3>, plump <lora:angelicv3-06:0.3>, long dark hair, undersized clothes, EpicRealism<lora:epiCRealLife:0.6> <lora:epiCRealismHelper:0.7>, Negative: epiCNegative, watermark, text, ((open mouth, teeth)), (((big boobs, cleavage, busty, nude, naked, nsfw,))) cartoon, painting, illustration, (worst quality, low quality, normal quality:2), mirrors, chinese, ((blurry)), fake, (((text))), (((watermark))), wrong anatomy, bad anatomy, extra limbs, ((bindi)) ADetailer: pakistani female, pretty, subtile make up, eyeliner wing, mascara, brown eyes I use 2M Karras or SDE Karras, 20 steps, CFG scale 5

>>24712 Not bad! Used alissv1-06 for this one; I like the little details it adds to skin. First image is the raw initial generation, and the second is after running my workflow on it. I'll definitely play around with the others soon and report back with my findings.

>>24713 BTW when I have a good seed I use the hires fix upscale

>>24717 Nice! :)

>>24712 Maaan, that's awesome, good work! Really great to see more folks pushing to improve the realism. The results you get with non-fat checkpoint look awesome. Using 'angelic' lora on a quite a semi-realistic checkpoint. Who's the actual name of the lady used for training btw? :D

>>24750 https:// coomer. su/onlyfans/user/angelicgiantess for the angelic Lora

>>24843 Nice. Thanks! Was not aware of her. And the 'curvy' LoRa - is it curvybabes ( https://bbw-chan.link/bbw/res/149634.html ) ?

>>24866 curvy = curvydelightxoxo

>>24880 thanks for info and thanks for share! your loras work very well, good job meng

>>24712 Any chance we could get a re-up? Would be greatly appreciated!

Could you generate some cute asian BBW girls?

>>24938 Here are some you might like: >>22279

>>24949 They are good, but they are too big for my tastes.

>>24951 What? I don't understand what you are trying to say?

>>24938 yes

>>24958 >>24959 I hope my future girlfriend looks like one of these girls. Could you make more please?

>>24938 Here are some chubby attempts. The first two still look more on the 2.5d side, but adding filmgrain and photograph to the prompts in the latter two really help with the realism

It is possible to make stretch marks on belly and small amount of cellulite too?

>>24979 These girls have cellulite. An excessive amount on some of them actually.

>>24959 The last pic is a top class, Sir! Would love to learn more about your workflow if you don'd mind. Do you use Auto1111 web UI or Comfy? Which model do you use? How do you add film grain?

>>24692 more arab girls plz

Reup of the Loras I trained: https://mab.to/t/zaM1jceKZe0/eu1 I trained one model with stretchmarks called yoki, this is an result of using that lora. I use automatic1111 and most of the time epicrealism model, then I upscale with hiresfix and put the picture in photoshop. I use dehancer to make it look like a little bit more like an actual photograph and put filmgrain on it aswell with that plugin

>>25068 Its nice could you generate more of that? does it can generate stretchmarks on the butt? Can it make round bellies too?

>>25068 Thank you so much! They work great! I'll post some of my outputs later on when I get the chance

Could definitely use more ssbbws

>>25091 Here's a few outputs from the past day, I've barely messed with it tbh. All inpainting results, I don't really do txt2img

>>25063 some imperfections but they're arab hehe

>>25205 this is purely halal, awesome work mate

>>25068 Any chance for a reup?

Bump

these are amazing, more arab girls plz bump

>>25068 +1 on re-up!

>>25068 It might be worth just upping it on Megaupload. Not like anyone's going to try to take it down. Or put it on civitai - that place is full of weird mutant sex dragons so why'd anyone go after something as simple..

>>25068 what the hell, that's excellent. Is anyone taking down the loras? Might be worth putting it up somewhere permanent.

>>25203 >>25205 make an OF and collect infinite amounts of oil shekels

how do you use loras? wtf are they??

If anyone is interested, I have a request. Can anyone generate a realistic sumo wrestling woman?

where should I reup the Loras?

>>25436 These are so good, do you think you could do some bigger sizes? Either SSBBW or like skinny upper half and fat lower half would be cool to see ; )

>>25436 Just put them on Civitai (https://civitai.com/). They make it super easy and that's where most people put there stuff. Another option is HuggingFace (https://huggingface.co/).

>>25436 wow, really impressive work. the first one is just pure magic >>25443 +1 put on civitai

I will look into civitai to upload the loras

>>24959 >>24958 Could you do some more east Asians please?

>>24692 Could you generate some chubby/fat Japanese girls with apple body shape? If possible making the pictures look like candids would be great as well.

:D

Dropping some generations of mine that I've enjoyed to keep the thread alive.

>>25508 Can we get a re-up please, these loras are an absolute gold mine and I accidentally deleted them 😭

>>25068 These are incredible results, please do re-up if you are able!

I'm gonna try and make my own aliss lora, more news later if it works out

>>25865 guys, this was surprisingly a success. will be uploading to civitai soon

just a thought, what other models deserve loras? thinking models with unique builds. might try taytay next

Cooked this one up today. not as chunky, but I still find it good

here are my loras for aliss, not OP btw just some models I trained by myself https://gofile.io/d/pNrKHg

>>25796 Nice stuff! Are you using Pony or SD1.5 based checkpoint?

>>24692 >>24708 >>24958 >>24959 >>25203 >>25205 >>25436 >>25437 >>25508 These are so realistic and I’m gonna lose it

>>26044 bruh have you seen a woman before

>>24692 These almost look like they have film grain. Is that intentional?

>>24712 Can we please have a re-up of the LORAs? Can't believe we had a master show up, show us some of the best AI fatties around and then disappear into the ether.

>>26268 Hey I am sorry to keep yall waiting, I had no fun creating these pictures anymore so I took a break, i don't know how but I can't get them as nice as before. The pictures im generating now just look fake and I dont know what causes this. But I can reup the loras maybe some of you can get nice pictures with them https://mab.to/t/jOaV30U54mb/eu1

>>26765 I mean, you've got the original images, look at the prompt, loras used. If you haven't updated any of your models, no way you can't get the same result once again. I use the same seed all the time, with identical models you're going to get identical pictures back. Do you use tweakPNG? small utility that reads png and allows you to look at the prompt that SD always embeds into the png..

>>26765 Could you please upload again? Didn't have access to my computer to check.

>>26774 I think i'ts kinda working again, I reinstalled SD and lost all settings and stuff also the original pngs

>>26905 I love how good your model is at generating belly rolls and hangs. I'm also very much into the double belly with extra fat lower belly aesthetic. I'm mostly using open art though and their models really struggle with that look. You either get s beach ball, a very wide yet somehow still flat belly, or an abomination made out of vertical flesh folds and several belly buttons 🤣

>>26905 these are gold man. what app do you use to do them?

>>27014 it's SD 1.5 with epicrealism and a bunch of loras. I tried it out, yeah, it kinda works but holy shit I forgot how bad 1.5 is with staying with the prompt. I feel like it's only good for inpainting, 95% of the time it makes some weird monstrosity unlike PonyXL which does it 15% of the time. Might be worth re-making those loras for PonyXL. How many images did OP use to make them ? It's not weeks of GPU going full tilt right?

>>27057 what about FLUX? have you tried it? it looks really impressive regarding realism and prompt adherence, so if you have some experience in training loras, why not giving a try FLUX.dev?

>>27060 1) it's not "flux.dev" you faget 2) I checked it out, it's nothing great. Certainly at generating sexy photos it's not better than pony xl, nor that great in prompt adherence.

PonyXL without any of these loras, with a skin realism and pregnant detailer lora. It's not quite right but the skin detail is not bad. Also extra pissed because gimpshop is being completely unreasonable and hiding tool options window.. so I can't fix the navel. Upscaling takes forever and it messed it up..

just leaving this one here for the little "faget" >>27091: https://civitai.com/models/683557/flux-bbw-women

>>27443 Great stuff dude, any adjustments to your model or prompt recently?

>>27550 thanks dude not really ;) I will post some more

more larger generations

>>27443 You should do more bbw Asian woman: Korean, Chinese, Japanese, Vietnamese

>>28384 those are all the same

ChatGPT estimated Bea to be aged 18, 163 cm, 130 kg, neck circumference 45-48 cm, bust 130 cm, waist 135 cm, hips 140 cm, WHR 0.96. It described Bea's body thus: "Body Type and Shape: • Apple-shaped: She carries most of her weight around her midsection, with a prominent, rounded belly that extends outward and hangs slightly, Her upper body (waist, chest) is fuller compared to her hips and legs, which are slightly narrower in proportion. • Voluminous: Her overall size is large, with a soft, abundant appearance in all areas, particularly her torso • Full-figured: Her body is generously proportioned, with fullness in her belly, chest, and arms. Round: Her midsection, particularly her belly, has a soft, circular shape that defines her apple-shaped figure • Bulging: Her belly appears prominent and bulges outward, giving her torso a spherical look. • Soft: The texture of her midsection and body seems soft and doughy. with visible layers of fat, • Thick: Her waist is wide and thick, with multiple folds and a large circumference that contrasts with her somewhat slimmer hips. • Heavyset: Her body exudes a sense of heaviness, esp. around her midsection. • Protruding: Her belly extends forward and appears to rest over her thighs when seated, creating a noticeable hang or belly apron. Pillowy: Her belly and arms have a soft, cushion-like quality, with a notable lack of definition, contributing to her overall roundness. Midsection Focus: • Layered: Her midsection shows multiple fat rolls or folds, particularly in her seated position, emphasizing the softness and abundance of fat in her belly, • Overhanging: Her belly hangs slightly over her waistband, creating a "belly apron" effect, typical of an apple-shaped body. • Wide: Her waist is broad and dominates her torso, creating a visual width that makes her midsection the most prominent feature of her body." Does Bea look realistic? Do you agree with ChatGPT assessment of her?

google's ai is surprisingly good at this stuff

>>28565 Great work. Could you please offer a prompt for those so that we can take a crack at the filter?

>>27442 Omg dude, please please share a guide/loras for us smooth-brainers on how to create images like yours. These are the best AI images I've ever seen.

>>28600 Heres one of the prompts I used real candid photos taken on bad quality camera. in the first photo there is s e x y black-haired pale american beautiful college-aged woman wearing a cute, sed uct ive witch halloween costume. she is pudgy and overweight, bellyfat hangs over waistband. her belly has gained weight. taking a mirror selfie. in the second photo it shows her side profile.

>>28646 Thanks man. Fun doing that one, and just writing in the "second image" that she's fatter. It does a good job keeping the AI person consistent between the two images.

Can someone bbwify hannahowk

I keep getting this on Google ai I'm still learning how to generate certain kinds of images, so I might not be able to create exactly what you're looking for yet. Also, I can't help with photorealistic images of identifiable people, children, or other images that go against my guidelines. If you'd like to ask for something else, just let me know! Any ideas how to fix

>>28791 Yeah, is there a trick to it? It seems to hate anything about weight/bellies when I give it prompts. I've tried Gemini and ImageFX in the test kitchen.

Google is so good

Not bad.

>>29295 how?

Yeah, Google does these very easy

>>29545 Does google have text to image? What prompt are you using tho?

>>29545 Send a link to the google ai

>>29545 >>29554 >>29557 Answering questions, Note I am not the original poster. He's using Google ImageFX, it's a beta but it let me right in, it also has a huge limit for daily prompts, which I can't find a set number for anywhere but it's over 50. When you prompt you can "edit" see the 4th image for an example. If you type in "overweight" or "obese" "big belly" it'll add that. you can also lock the seed and set the prompt in the "edit" box to be the clothing description from the original prompt box in order to generate a fat version. Here's a quick example. Note I'm not the other guy, he seems to be doing a before and after type prompt, but my method works so whatever.

>>29554 >>29557 I'm that poster. Yeah, it's a before and after, I just used this post >>28656 as a template.

realistic gens

.

.

>>29628 did FX remove the "edit" button? I signed up a couple of days ago and it's not showing up for me, I can only create new images (or "remix" them, which looks like just taking the prompt for the image and messing with it)

Hallo Anons, I am happy to have found this thread. I am an artist / photographer who has created art for over 30 years and new to AI. My work typically contrasts the extreme anatomical differences of everyday people in everyday situations, usually women. Much my work has highlighted obesity and body positivity. AI has seemed like an amazing artistic tool - I often have concepts for portraits but can’t find real-life models or subjects to capture the idea. My key problem is that I am looking for an easy to use image generator without content limitations or censorship. Generators like Openart and others I have tested seem to consider my images & prompts “NSFW” - although they clearly are not. I am NOT interested in creating NSFW porn images - I DO want the freedom to generate realistic images of my choosing - the images could be provocative and perhaps contain some nudity - or not. Attached are 4 images made with Google ImageFX that are very close to what I want to produce. ImageFX seems great limits the amount of images I can make - Is there similar commercial alternative?

>>30274 If you want generation freedom except for certain nsfw themes and celebrities then civit If you want full generation freedom then download a local generator

BBW boobs

busty bbw

great result

google labs is amazing for photorealistic

>>30751 Prompts?

>>30751 that fourth pic looks absolutely incredible. what prompts do you use?

prompt: before after comparison mirror selfie of 80lbs weight gain in 2 years of female Iraqi college student , in the before picture she is extremely skinny and fit, in the after picture she is extremely obese and overweight with big fat rolls , she wears natural make up , eyeliner wing , nose piercing , obese, overweight, wearing outgrown undersized clothes that do reveal her extreme fat rolls hanging out, muffin top , thick, cellulite, bad phone quality photo, snapchat , holiday photo , instagram post , three quarter side view, full body shot , she is facing the same direction in both pictures

someone mentioned that adding film grain helps add to the realism and I think that's the ticket. all of these generated and upscaled locally using the cyberrealistic pony model, I think they may even look as good as flux

Thanks to iraqi prompt guy got to know Imagefx, been playing with it, its hard to work with it but life changer indeed xD

>>31759 Goodjob man im that guy you referring to I guess haha, do you mind sharing the prompt you used?

>>31759 >>31769 Wow, really nice stuff both of you. Can we get prompts for those? They look great.

>>30860 this guy myy savior xD >>31769 Not sure you the same guy, these are unachieveable (bellies/tets/4ss) on Imgfx Note i couldnt master the arts of prompting in Imgfx, but i try my best and sometimes it works: (aspect ratio is important) seed: 226510 2/3 face low angle before after comparison photo, of 900lbs massive weight gain in 2 years of female iraqi college student, wearing white sports bra and black sports shorts, she wears natural makeup , eyeliner wing. in the before she is at a crowded gym, crawling diagonally, thin body smiling, braids hair. in the after picture she is crawling diagonally, at her room with toys, obese body, heavy-bottom body, obese face, double-chin, muffin-top, belly hang, knees on the floor, obese legs, cellulite, obese arms, cute smile, ponytail 822497 before after HD selfie comparison of 700lb massive weight gain of female university student iraqi, wearing black bandeau thin bikini, posing in the same room, full body shot. in the before picture she wears natural makeup and eyeliner wing, is smiling, extremely skinny body, anorexia body, has little chest. in the after picture she wears natural makeup and eyeliner wing, has quadrupled her weight, she is a lot bigger now, is morbidly obese, morbidly obese arms, obese forearms, obese face, one hand rests in her waist while other hand gently grabs her hair, thick cellulite thighs and fat calves and huge legs, obese ankles, morbidly obese cellulite belly, facing the same direction in both pictures. smiling, three-quarter view 797264 before after comparison full shot of 200lb massive weight gain of female iraqi university student, wears white bandeau bikini, natural makeup and eyeliner wing, short ponytail, at the hospital. in the before picture she is smiling, connected to saline bag, extremely skinny body, skinny legs, holding a tiny blank marker white board, standing next to chubby female doctor. in the after full shot she is smiling, morbidly obese, has morbidly obese arms, heavy belly bottom, obese thighs, fat calves, obese legs. a doctor stands at her side, holds another tiny blank marker white board

Ai is soo good

anyone have any tips to improve face?

had some success with comfy ui and "leaked_nudes" lora

>>32965 Hot damn, you've still got it.

>>32972 thanks hehe, kling is insane btw and this is only the 1.6 model: https://gofile.io/d/5FdCAk

Natali conti

>>32961 the background...

What are y’all using to make these crazy photos?

>>32974 Any tips with prompting kling for videos?

>>33072 AI

Some of my creations on Sora.

>>33243 amazing what do you use to generate?

Gemini experimental

>>33243 >Gemini experimental That's really awesome. Can you share your prompt?

>>33236 prompt?

>>33243 this is peak, do you have any more of your work you can share?

My humble contribution, how do you y’all get get Gemini to pump out something so explicit?

Took a random image from this thread to try out the new LTX video generation model online via LTX Studio. Turns out well, I think I attached the image I used, thanks to the initial creator who uploaded it. I cannot upload the video here, so gofile: https://gofile.io/d/cEFqma The nice thing is, if you have a beefy enough machine, you can run this videogenerator locally. Here an installation guide (this was the video which led me to LTX in the first time): https://www.youtube.com/watch?v=D56ZHGvk9ks if someone has success with this, make a thread, post results, I am curious to see what you are doing with this...

Was able to make this with chatgpt. Is there a way I can make an ai turn it into a video?

I'm not familiar with a.i at all but if someone is feeling generous could they animate this:

>>33072 unell stated otherwise assume its someone with a decent NVIDIA graphics card using comfyUI with wathever model that is popping off on civitAI. If you want to get started just search "comfyUI tutorial on youtube and play with the prompts

>>33832 Choke her during missionary. And hold her thigh up into her chest tell her who daddy is.

goated workflow

i'll some that i think are pretty nice

>>30860 followed your instructions and managed to get imagefx/whisk to make some very insanely realistic pictures lol

Can somebody make a realistic version of this sumo geisha girl?

Insane.

model: Chroma

seed: 425306 Prompt: 4k photo in lighting which reveals all blemishes, cuts, bruises, cellulite and other imperfections. Cutie cute brown-eyed, fat cankles obese immoble 600lbs uk uni model in a white whirlwl that makes people fat. Her makeup is light-touch but stunning. her skin and face is dimpled and imperfect The photo is sideways sideways sideways lays on her side in a obese photo on a bed. Her hands and wrists are very fat. Her ankles and feet are rolly fat. she has turned into a melted pile of the fattest woman in the world, immobile fat. she smiles confidently. she has no muscle definition, eats bad food, and clearly never exercises. a feeding tube is in her mouth. Her tummy is dimpled and doughy and covered in rough cellulite, painted with the phrase 'lazy blob'. Her neck and chin fat hang down touching her tiny breasts huge puffy auriola. Her body looks like a melting pile of dough. her fupa is shaved. Her arm fat rolls over her elbows, falling halfway down her forearms. Her arms are tied up from the ceiling, and forearm fat rolls onto her hands. Her thighs have multiple rolls of flab fighting for space, and spilling down. Her treetrunk calves roll, so fat that it that pools, and rolls onto her feet like cankles. we can only see half her feet because of the fat; her toes are, and are inflated and puffy, and her feet stretch and point out. Her hands are inflated and rest on her fupa. Her skin is marbelled with cellulite and stretch marks. She is as wide as a bed. a panties lay on the floor.

>>34519 >can someone make her fat no fuck you, do it yourself. picrel is a fresh gen, get mogged

>>34522 also, any tips on making hands better? no lora I tried seems to make a massive difference

get some detox from the 500 pound beasts

If youre a lazy fuck (like me), make chat gpt write out prompts for image generation. Surprisingly, the results of a chat gpt prompt are generally better then my home cooked ones, I just add details to tweak it.

Some nice one's

A few I recently liked

These are great, what did you use to make them and what prompt did you use?

Here are some i created with Gemini

Why not upload these at once? Its wasting my mouse wheel scrolling though >>34925 >>34926 >>34927

Its that time of year, who doesn't love a trip to the State Fair

... and for anyone who hasnt been to an American state fair in the midwest, this is exactly what you see.

>>35126 >>35127 These are amazing, what kinds of prompts did you use?

>>35132 My approach is essentially descriptive writing, Concise descriptions of the place, what the person is wearing or doing, and describes the person. I find if you describe how the clothing fits you get great "clothed" results - If you describe the body you'll get more skin. On top of that, I described a professional artistic portrait photo-essay by "famous photographer' and "well composed" ... that kinda stuff. By building the prompt in modular sections you can fine-tune what you want - keep the location, change the subject - or keep the subject / change the location etc etc etc. I dont save my prompts

>>35126 how do you manage to achieve corsets on fat ppl?, i just cant get it to work

>>35126 >>35127 Oh my god this needs to be a series, especially the woman in blue and the one sitting down in white

>>35142 its been hit and miss on corsets - but describing the corset / clothes in detail and being more simple about the subject's obesity seems to make the generator think harder about the clothing. completely rewriting prompts can drastically change the result of very similar descriptions - I dont save the prompts because I'm looking for variety and want different people, tapping out a fresh description does that. ^^ On the blonde in the corset image I do remember adding a "Emphasis on abnormally blubbery fat arms with creases... " etc etc etc at the very end.

These are pretty sick ngl.

... A few more fat fair girls.

this batch with a fatter version of the girl on break. Was going for super fat but believably. Anyone ever able to get "believably fat" but with stomach blubber that hangs below the knees when standing? ... or even sagging onto the ground - inside a pair of leggings or something? At a certain point ImageFx just turns out run-of-the-mill ai blobs that are huge and less real

>>35161 Iveen trying comfyui and i dont seem to be able to make remarkably fat bitches in it, tried Modesl like cyberrealism pony, pony realism, illustrous, and now realdream tried different loras, like: https://civitai.com/models/675260/extreme-obesity-ssbbw-generator-for-females and https://civitai.com/models/461046/immobile-extreme-fat-pony-xl?tags=3915 fuk am i doing wrong? i dont even want blobs, just like, 600lbrs or something

>>35162 we need to train a model (lora), theres a decent one on Flux but Flux is not realistic enough for me. I have trouble gathering a dataset for WAN 2.2 lora's because models are so lazy and don't use good photo quality.

>>35165 unrelated, but I can tell that you arent filtering masculine features. In the negative prompt, include anything masculine to fix this. Also, you should try https://civitai.com/models/277058?modelVersionId=1522905 I have not experimented much with ssbbws, they aren't my thing, but they are possible, although the quality and stability drops the larger you are trying to make a woman. Provided are examples of how sixe correlates to realism

Been playing around with WAN

>>35225 Got somewhat decent results with https://civitai.com/models/2664/bigger-girls-models?modelVersionId=56154 though this has adetailer and the face is still fucked up. the model is not bad but i feel you have to be more specific than usual as to not get glitches. I still dont know how to do the face-swap thing nor complex comfyui setups so might be screwed for the time being. Last one was made with epicrealism, its k

>>35258 >I still dont know how to do the face-swap thing if you dont feel like setting up the workflow, there are some to download. try out: https://github.com/amao2001/ganloss-latent-space/blob/main/workflow/2025-02-10%20ace%20reference%20editing.png

sharing prompt cuz gatekeeping its just dumb 373558 natural colors, daytime, sunrays, before after comparison full body shots of impossibly weight gain in 5 years of female armenian college student, tanned nose, she wears natural makeup, eyeliner wing, fake eyelashes, blonde ponytail, brown eyes, freckles, pierced navel, tiny black twisted bandeau bikini, sideview at her kitchen, with lots of enormous jars with weight gain and body mass suplements and shakes on top of the kitchen counter. In the before picture she is extremely skinny and fit, held a tiny white marker board with the written sentence "fat scale challenge" with food doodles, softly hand on her waist, happy. in the after picture she has expanded into a gigantically wide mountain taking up the entire room as her waves of fat spill outward in all directions, is impossibly obese, she is rapidly growing fatter, she is beyond the fattest woman in the universe, magic weight gain has caused her to swell to impossible fatness, every limb is extremely swollen with fat, particular emphasis on her gigantic fluffy body, obese cheeks, holds a big funnel connected to a thick garden hose, hand on her hair, proud of her gain, shy smile. sideways. at the back festive letters balloons "500!",

383619 natural colors, daytime, sunrays, blury background, before after comparison full body photos of impossibly weight gain in 5 years of female argentinian college student, extremly tanned skin, she wears natural makeup, eyeliner wing, fake eyelashes, blonde ponytail, brown eyes, freckles, neck choker, black twisted bandeau bikini, sideview at her kitchen, with lots of enormous jars with weight gain and body mass suplements and shakes on top of the kitchen counter, tiny white marker board with the written sentence "road to ssbbw" with food doodles at both photos at the same spot. In the before picture she is perky, softly hand on her waist, happy. in the after picture her bikini doesnt fit, her skin is marbelled with skin folds, she has expanded into a gigantically wide mountain of fat and creases spill outward in all directions, is impossibly obese, she is rapidly growing fatter, she is beyond the fattest woman in the universe, magic weight gain has caused her to grow to impossible fatness, every limb is extremely obese, particular emphasis on her fluffy obese arms with creases, her layered fat rolls ressembling Michelin mascot, obese cheeks, happy pleasure. shrugging one shoulder, one hand on her waist, measuring tape, poses sideways, blender jar with tons of butters.

>>35347 "only 500" pathetic

>>35355 Lol but my idea is to make it so its believable.Realistic as thread says xD. im happy you played with that prompt. its a mix of all that has been said/shared here if u read carefully

you can drag png into comfyui for workflow

>>35450 >>35449 >>35448 >>35447 holy hell, what model/workflow do you use?

>>35447 This isn't true I don't think. At least I can't open these images as Comfy workflows. Maybe the chan strips the metadata from the uploaded images? Would definitely _really_ love to see your workflow.

>>35469 I take it back, was able to load it. Great stuff, thanks!

>>35470 Great! Glad to hear it worked im curious to see others results too! Image to Video works great as well

>>25945 Reup please?

Been testing out a new lora I made.

>>36204 I have 2 versions of the lora that I trained on qwen and SDXL. The SDXL one I've been using with Pony Diffusion 2.23, but it seems like it's harder to get naturally realistic elements. It still looks painterly in parts of them. But if I use the qwen lora with the qwen checkpoint, then things start to look much more real.

>>36206 great work mind sharing your lora?

>>36209 Definitely! I'll be dropping it on Civit AI once I know it's good enough. I was trying to parse through what you were doing to make you image and video gens, but instead, would you mind making a consolidated post of your checkpoints and loras?

>>36209 Here ya go https://civitai.com/models/2038661?modelVersionId=2307398

>>36217 thanks man, do you mind sharing your dataset? I can train it for Flux 1 dev (krea) if you want :)

>>36267 Sure, but the dataset isn't the greatest. This was more of a test to see how a large dataset worked with not great captions. The captions would need a lot of cleanup and there are over 1800 images with them. The captions were auto generated for me with a few key words added. I manually cleaned up only a handful of them. https://gofile.io/d/csc6C9 However, I did make a wildcard with a huge number of prompts already in it, so you can just run a long generation and get a nice big variety without having to baby it. https://justpaste.it/8u0iq