/bbwai/ - bbwai

Stable Diffusion, Loras, Chatbots

BBW-Chan is supported by simple text or static image ads from a-ads.com.
Please consider whitelisting us in your adblocker if you enjoy your stay!


Mode: Reply
Name
Subject
Message

Max message length: 9999

Files

Max file size: 10.00 MB

Max files: 6

Captcha
E-mail
Password

(used to delete files and postings)

Misc

READ THE *RULES* BEFORE YOU POST!

[ / / ]

(953.28 KB 1024x1024 00027-2179109774.png)
(1.10 MB 1024x1024 00229-4084635802.png)
(952.67 KB 1024x1024 00327-4084635703.png)
(1.06 MB 1024x1024 00067-2179109820.png)
(1.15 MB 1024x1024 00096-4084635666.png)
White USSBBW Seneca 09/04/2023 (Mon) 12:35:27 No. 12302
Some USSBBW. Ai still has a long way to go to perfection. And what did you find or create?
>>12302 These are incredible! Where did you find these/did you prompt them?
(1.20 MB 1024x1024 01.png)
(1.24 MB 1024x1024 02.png)
(1.33 MB 1024x1024 04.png)
(1.32 MB 1024x1024 03.png)
These could probably be improved a lot with inpainting etc if I wasn't so lazy
Great work!
These are amazing! Which ai program did you use, and which prompt?
(354.20 KB 512x512 CutieFace.png)
Bamp!
(314.76 KB 512x512 BellyFromBehind.png)
This will change the culture here for sure...
Need more of this but I need to get set up. I would like to use my laptop if at all possible.
Uuhhh damn. These sure are something. More? Like a bunch more?
(2.74 MB 3072x3072 IMG_0983.jpeg)
>>12384 Can’t she ded
(1.66 MB 1199x1199 PinkWave.png)
(980.53 KB 1022x646 blorb.png)
(1.85 MB 1128x1344 FieldImmobility.png)
(925.77 KB 896x768 Maid.png)
(1.80 MB 1404x1020 CouchSitting.png)
(1.71 MB 1259x1260 Leaking.png)
I don't often do USSBBW/Immobile, but here's a few I've done over the past few months. (last one spoilered for lactation and not being 'immobile' per se)
>>12407 I’d love to know if you can do an immobile pear shape, where she has a noticeably thinner torso and massive lower body, with her legs in almost a complete split?
>>12407 Holy shit, the first and second are amazing! Can you make more like the first two?
I got to know what AI you're using, op.
>>12410 OP is using Stable Diffusion 1.5 You can get the firefox plugin that will tell you the prompt The lora is really good. I recommend it.
(106.55 KB 1024x1024 ComeHaveASeatBabeImFinished.jpg)
Anyone here is doing God's work!
(137.01 KB 1024x1024 ToiletGirl.jpg)
>>12415 Bumperino!
(695.46 KB 950x839 disc101.png)
(662.73 KB 1067x600 Echo1.png)
(832.37 KB 1083x611 Echo2.png)
(916.99 KB 794x900 Echo3.png)
>>12408 not exactly what you're asking, but the first image is from a request I had for something similar. Their reference image included the extremely waist. >>12409 again, not exactly what you asked, but I did a few immobile/blob morphs of Echo from another request which heavily relied on SD to create.
>>12407 Could you give us the metadata for the 2nd and 3rd images? Those two are absolutely insane
>>12379 I'd love to see more of this. I love everything about it, the facial expressions, how the face is just burried in fat.
(1.04 MB 1024x1024 LilHelpBabe.png)
>>12318 These are really good. I would love the details. l fought with this image for a while, but it wouldn't converge. I used the Lora that OP mentioned: BGV5EX, which has "immobile" and "blob". This Lora is pretty good: https://civitai.com/models/48197/obese-girls-or-concept it has weightimmobile .
(94.52 KB 1024x1024 MakingRoomForMore.jpg)
>>12475 Thanks! I think that the best ones have subtlety.
>>12503 Greta Thunberg has let herself go, lol.
>>12500 Is this the new Primus album?
These are amazing, what prompt are you using? Thanks in advance
>12515 There is a mix of us in here. Stable Diffusion skills are hard fought! And even if you are good, it is hit and miss.
(116.50 KB 1024x1024 LilHelpBabe.jpg)
>>12506 I was able to do a repair job.
(350.76 KB 480x512 00026-2953762361.png)
(738.57 KB 688x728 00377-2902862161.png)
>>12433 yes and no. For 'blorb', I remember I started with a reference image, but I can't find it again. I know it was a drawing, but I didn't save it. Apologies. For Field Immobility, I've attached the first generated image I used as a basis for it. you should be able to pull the metadata from it. The reason the images I share here don't have metadata attached is because I take a layered approach to creating these. As you can see, the stating image was done using a drawn-style checkpoint, and then after touching it up in photoshop I sent it to img2img to upscale and convert to a 'realistic' style (EpicRealismV3 in this case). You can see the first result in the following image. Then it's a process of touching the image up in PS, feeding that back into SD, and repeating the process a few times. I currently can't get inpainting to work at all (possibly due to the fact that I'm running AMD hardware), so I have to crop portions of a semi-finalized image and reprocess them to get cleaner results (most often faces and hands). Then I collage the bits together and export the final image.
(2.24 MB 1380x1380 HeavyElevator.png)
(2.18 MB 1428x1428 HeavyCafe.png)
(1.17 MB 1032x1032 OutdoorMall2.png)
(2.47 MB 1488x1488 LilHelpBabe2.png)
For those of you who don't care how the sausage is made, here's a couple more USSBBWs. >>12500 >>12520 I also took a crack at it.
>>12526 Love a good belly dragger. Can you make some with massive boobs to match?
>>12526 Holy shit yes
(2.71 MB 1440x1440 HeavyFullDen.png)
(2.75 MB 1512x1443 heavyCouch2.png)
(2.45 MB 1392x1392 FullPark2.png)
>>12530 WHY NOT. Even threw in some hyperbreasts because work was boring today.
>>12543 Dude, you’re KILLING IT!!! Absolutely magnificent work! Please DON’T stop!!!
>>12543 Your hyper girls are seriously some of the best I’ve ever seen. Especially the hyper asses
(2.21 MB 1524x1200 WideStage2.png)
>>12408 I didn't forget about you, friend. >>12544 >>12554 >>12542 Thank you all kindly. But I'm just an asshole with a copy of photoshop, an install of SD, and too much free time. I haven't done anything you all couldn't also do.
>>12602 You're stuff is really good. Thanks for uploading
>>12602 Love it when they just melt into a puddle of ASS. Can you do one this style, but lying flat?
(2.10 MB 1440x1149 Widebed.png)
>>12613 You mean something like this?
>>12621 I meant lying on her back tbh, but this is amazing
>>12602 I love this! It’d be great to see some more like it, with her being skinny and slim but from her breasts down she’s an immobile blob~
man this is going to create so much more cursed fetishfuel that's going to turn the next generation to perverts
(1.91 MB 1212x1500 HeavyDoor.png)
>>12664 In the words of everyone's favorite light-transport researcher: what a time to be alive!
>>12717 More of this please! Loving all the massive fatties here regardless, but I need some distressed ones to hit that swelling horror vibe!
Absolutely amazing work, please don't stop ever! Would love to see a massive girl lay on a bed, her huge legs as spread apart as much as she can manage, holding back her belly to expose her buried pussy crack, glistening wet and ready for a pounding! 🤤
>>12717 I'd like to see this one from behind, and stuff like this where instead of a shocked look on her face, she has a seductive look - like she knows you love huge women and she's coming into the room to fuck your brains out.
>>12543 SPECTACULAR!!!
My first attempt, just playing around a bit, hope you approve
(1.00 MB 1024x1024 00000-296500888.png)
(966.08 KB 1024x1024 00055-296500943.png)
(1.16 MB 1024x1024 00018-296500906.png)
(1.14 MB 1024x1024 00025-296500913.png)
I don't know why asking for green eyes gave me green veins
>>13322 Are you using stable diffusion on the PC? I've tried only with online generators and not got anywhere neer this level or realism.
(3.32 MB 1368x1368 NewFirst.png)
(2.34 MB 1344x1310 Greeneyes.png)
>>13189 Gave her a bit of an uplift. >>13322 Tried cleaning one of these up for you.
>>13354 can you make some of this ones into a realistic version ?
(857.06 KB 824x1032 00062-3129837506.png)
>>12525 Thanks for sharing your methods its been really usefull!, please do tell how you manage to add that fine skin detail for two ladies here>>13354
>>13330 I'm using SD on my machine. I followed this guide here: https://rentry.org/voldy#offical-stable-diffusion With this checkpoint: https://civitai.com/models/25694/epicrealism And these loras: https://civitai.com/models/40468/cellulite https://civitai.com/models/112552/weight-slider-lora https://civitai.com/models/111635?modelVersionId=120467
>>13354 Thanks for cleaning up the image on the right. How do you go about doing that. I tried inpainting but I can't get the hang on it.
(611.83 KB 896x976 Himeno1.png)
(2.03 MB 1536x1536 Seneca1.png)
>>13366 SoftestCore is a very tall order as their work is heavily stylized and has truly impossible dimensions. They're a fantastic artist, and I don't think I'll ever be able to convert their stuff into a realistic style. I think that's for the best, anyway. The simpler Himeno/Power ones are a bit easier, but still tough. Here's what I have now, but it's not the best. >>13374 >>13407 I think I can answer you both at the same time. Skin details just come down to a combination of the right LoRAs and checkpoints, I rely on NextPhoto and the previously mentioned Cellulite LoRA. More broadly, I 'cleaned up' that one by simply reprocessing the image with NextPhoto, BiggerGirlsModelREAL, and lots of photoshop tweaking. Since all four images you shared had a significant green cast, you might want to check what VAE you're using, if you're using one at all. I don't know much about VAEs, but I can generally recommend vae-ft-mse-840000-ema-pruned.safetensors. It's technically for anime (or so I'm told), but I've had pretty good luck with it for realistic images. Sometimes things get a little oversaturated, but nothing a quick photoshop tweak can't fix.
>>13374 Now THIS is amazing!
>>13432 can you do more like the first one ? and also it would be awesome if you make some ussbbw using their bellies as tables
(9.70 MB 2608x2765 The Lap of Luxury.png)
(134.22 KB 900x570 dream1.jpg)
(118.50 KB 900x514 dream2.jpg)
(279.08 KB 1600x822 dream3.jpg)
(262.70 KB 1600x823 dream4.jpg)
>>12543 >>12621 >>12717 >>13432 These are astounding. Any chance to make these ultra realistic?
(2.09 MB 1373x1376 SoftestCore1.png)
>>13432 >>13486 apparently I am a liar. >>13488 I was gonna say probably not, but after getting the result I did on this one, I guess I should say maybe.
(2.50 MB 1800x926 dream3.png)
>>13488 With some more experimentation, this is where I got. A few notes: > Absurdly fat faces (like ones with gigantic cheeks) are nearly impossible for me. I've done puffy, multiple chins, round, etc, but this specific thing (huge sagging cheeks) is just beyond me at this point. > Huge puffy hands are pretty much the same. SD (and most other models) really struggle with hands when they're 'normal', deformed/unusual hands just devolve into flesh nightmares or disappear completely. > converting existing artwork into 'realism' is more work than engaging for me. Not sure what the specific difference is, but I just don't seem to enjoy the process as much as when I'm generating something whole-cloth, or when I'm boosting my morph work.
>>13637 holy fuck this is beyond hot, i see what you mean with the difficulty but this is absurdly good! i cannot thank you enough for entertaining my request. hopefully ai will evolve enough to make tons of stuff like this with ease
>>13630 maybe ...some day
>>13630 and ...with just a little of photoshop
>>13637 muh dick, do more like this
>>13637 It's like a really fucked up sitcom.
super fat face
>>13637 please do more like this! maybe with the second page of the comic
Fucking slam dunk, it’s like, where I always hoped Kraban could get to- awesome work
(3.15 MB 1254x1584 HeavyParking3.png)
(3.35 MB 1296x1958 HeavyVenice1.png)
(3.04 MB 1424x1824 HeavyParking1.png)
can someone make more of these ? ...and also bigger
Any chance that the Recipients 3 images could be tackled with AI wizardry?
(550.44 KB 1800x1286 IMG_20230323_150834.jpg)
Could someone do a hiperrealistic version of this one?
(4.36 MB 1866x1472 HeavyOffice2.png)
(3.20 MB 1308x1632 HeavyParking2.png)
>>13488 >dream1.jpg) Is this AI generated? If yes it is another proof that AI is nothing more then art theft. I recognize these frames and they look like from that old ass comic i did read years ago.
(649.60 KB 552x697 HeavyOffice1.png)
>>14383 My guy, please take a breath. The person who posted those was asking for someone to try and recreate them in a realistic style. You're on the right track though, is is an old ass comic (fuck I remember when they were fresh). They're from StudioFA, a series called Kitchen Bed. You can find the originals here: https://www.deviantart.com/studiofa/art/Kitchen-Bed-210480314 Additionally, this thread is explicitly for 'photorealistic' renderings. If you see any art here, it's most likely going to be someone asking for it to be rendered in this style. For everyone else: here's an image that's somewhat related: I was trying to recapture that early 2000's feel of tiny, poorly lit digicam photos. Y'know, the only thing we had back then.
>>14383 what a retard, it's like you only exist to discredit your own cause
>>14383 Faggot
(179.75 KB 1392x888 Capture.JPG)
Came across this the other day but wasn't able to recreate her shape. Can someone give it a shot?
(1.56 MB 1176x864 Saxon1.png)
(2.59 MB 1428x1428 HeavyInflation2.png)
>>14450 It's possible, but damn is it a lot of work. I could clean this up further, but I'm just not inclined. Much like the one I did of a SoftestCore piece here >>13630, I recommend folks just enjoy their art as is. >>13699 Hey! nice edit, well done! and here's an inflation inspired woman.
>>14498 I´m glad you like the edit, also can you make more like this two and also bigger ?
>>14498 That's awesome, once again your inflation stuff is top tier!
(1.88 MB 1265x994 HeavyBride1.png)
>>14554 can you make the table disappear and that they only use her belly as a table ?
First, your work is just incredible, so thanks for that!!! Second, please ignore my lousy drawing as I am NOT an artist - the picture proves that, but is it possible for you to create something like this pose? If not, thanks for your time and again, the incredible images you’ve already created.
(1.92 MB 1310x1030 HeavyBride2.png)
(565.95 KB 912x664 01655-4111812670.png)
(2.99 MB 1518x1008 HeavyOffice3.png)
>>14627 Jesus, enough is never enough for you gremlins, is it? See attached. >>14630 Thank you for the kind words! And don't feel bad, you at least put something on paper which is a far cry from a lot of people. As you your request, I'm afraid the answer is no. I've attached the best thing I was able to get out of it, and it's just... not good at all. I tried to salvage it, but it just kept descending into chaos. This is a perfect example why Image AI isn't going to replace anyone anytime soon. It truly can't handle 'novel' concepts. Until someone puts together a large dataset of fat women lying on their backs looking up at the camera, the tool just doesn't know what to do with it. That's also why the image is upside down; my goal was to get something presentable at a different angle and then fake it to look like she was lying down. Trust me, you don't want to see what it spit out in the normal orientation. Conversely, if I had an image of a woman in this pose (but with normal proportions), I could probably trick the AI into giving me what I want. In penance, here's an upscale of one of the office workers from >>14340.
>>14712 Loving your work dude, I'm following you on devart (different handle) What did you use to upscale the image?
>>14712 Also I get what you're saying about novel images; the more I Learn about AI images the more I realise it's effectively just a petern recognition system. Still very interesting though and who knows where it could lead. I remember early 2000s when voice recognition software started coming in and it was aweful whereas now alexa siri etc work pretty well.
>>14712 Oh hey, thanks! Hit me up on DA, I'd like to see your other work if you have any. As for my process, I'm going to do a full write-up of my process soon, I'll share it here when I publish it. But suffice to say, it's a blend of NextPhoto, BiggerGirlsModelREAL, a handful of different fat LoRAs (mainly Symix, cellulite, and ObeseGirls) and constant tweaking in photoshop. I iterate by increasing the image size by 10-15% with each pass, until one of my dimensions is greater than 1000. I take that back into PS, upscale with Preserve Details 2.0, then select parts of the image to reprocess (mainly hands, feet, and faces, but also changing size and shape of individual elements if needed) and layer those results back into the final image. From there it's color correction, detail cleaning, and a final sharp pass. Basically, I treat it like I would an actual photo. >>14734 God, I was working at CompUSA when Dragon Naturally Speaking 8 came out, it was revolutionary. Now I get mildly irritated when my car stereo slightly mishears my request.
>>14712 Yo, thanks for the attempt! I get what you’re saying. This and maybe one pic of JJ is the closest I’ve ever seen that’s kinda close to what I attempted to draw. Also follow you on DA. Thanks again
>>14736 I've not really put anything on DA but I might upload some. I'm currently trying to get my laptop to run it using onboard GFX or alternatively use google colab or something. I'm not officially trained in anything so I'm basically just copying and pasting code most of the time. I would like to work as a coder but it's way off for me and my current job pays ok, so I'd probably end up taking a pay cut to get experience which sucks.
>>14712 Here’s one of the pics of JJ I was in reference to. The pic had her leaning more to the side. I straightened it but not sure any space you needed, so left the white in. Legs don’t need to be in the pic, I’d rather see a mountain of a belly. Now I’m being an asshole ASSUMING you’ll attempt something. If you can great, if not, you’ve done more than enough to quench my thirst. 🍺 Cheers!
(2.12 MB 1341x1079 HeavyBed2.png)
>>14739 >>14762 >A mountain of belly Not my best work, I'll admit, but I think this is what you were looking for.
>>14780 ABSOLUTELY PERFECT!!! THANK YOU SO MUCH!!!
(2.24 MB 1222x1229 RoundStudio1.png)
(3.26 MB 1560x1164 HeavyTile1.png)
>>12407 Wow, those megatit ones!
>>15390 That skin is so incredibly real
(2.76 MB 1070x1595 HeavyHallway1.png)
(1.35 MB 1200x1200 Bacchus.png)
>>15396 >>15397 Thank you kindly! It takes a decent amount of work to get it to look like that. and I'm still learning things all the time! Look at the first photo (made today) compared to the second. That's from only a little over a month ago.
>>15425 I’ve been beating my d!ck like it’s owed me money since you started posting your incredible work!!!
(312.32 KB 2782x3314 1698859644917.jpg)
Dude, your work is incredible. I wanted to as if you would give something from Huurin Kazan a try.
(617.31 KB 839x1000 HuurinKazan1.png)
>>15503 I'm generally not a fan of converting extant art to realism. Sometimes it's a little fun, but most of the time I prefer to leave art as it is. That being said, anything with those hyperfat cheeks and massive waddles generally don't resolve as there's next to no examples of that in real life. So, here's something 'inspired' by the piece you shared.
>>15614 Dude that's fantastic! Thank you!
Dude that's awesome! Thank you!
(2.09 MB 971x1452 HeavyMall1.png)
>>15390 perfect work again, i dont think i could nut any harder. the blobs you generate are legendary
>>15684 Outstanding! Think you could turn Jackie into the mega blob we’re all hoping she becomes?
(2.11 MB 1104x1260 JuicyJackie1.png)
(1.32 MB 890x1400 ThickHungryHoney1.png)
(3.00 MB 1482x1589 UdderlyAdorable5.png)
>>15767 yup. Have a ThickHungryHoney and an UdderlyAdorable while we're at it.
>>15771 You’ve outdone yourself with that JJ pic! You’re the Rembrandt of fat babes!!! THANKS!!!
>>15614 the triple chin of my wildest fantasies. anything with more huge faces pls
Would anyone be able to whip up a couple hyper bottom heavy images? Where the girl's upper body is only a little pudgy, but their lower body is essentially a lake of ass and thigh meat.
(3.00 MB 1224x1524 WideBeach2.png)
(3.27 MB 1291x1626 WideBeach3.png)
(3.53 MB 1433x1684 WideBeach1.png)
>>15994 THUNDEROUS
>>15771 How?!
>>16066 They gonna cause a tsunami if they start clapping those cheeks on the beach.
>>16066 A bit more belly and shorter ginger hair and freckles, these would be pretty much what I'm trying to get out of stuffer.ai but failing because it seems like keywords relating to hips make them huge all over and keywords relating to butts make them face away from the viewer...
(2.26 MB 1080x1620 WideBeach4.png)
(2.23 MB 1741x1192 WideFullStudio1.png)
(4.31 MB 2056x1352 WideMall4.png)
>>16099 Everyone loves a ginger.
(3.21 MB 1892x963 HeavyFastFood1.png)
let me put a little emphasis on the USS in USSBBW.
>>16066 If only the edges weren’t cropped off the first two :(
>>16152 Think you could do something with a shape of this woman, without the others in the pic? Doesn’t need to be dressed like her, or dressed at all for that matter. Her body type is magnificent!
>>16152 oh my fucking god more like this forever PLEASE this is so good
Just curious, has anyone ever tried to gen any mega blobs like pic related?
>>16066 It's only a matter of time before AI will do animations, and I can't wait until we can see TGWDAT's creations walking around or growing from a normal size to the epic proportions he is so adept at creating.
(4.07 MB 1293x1805 REquest1.png)
>>16167 How's this? >>16164 Oh, I don't know, I kinda like when a woman is too wide to be fully captured in frame.
(633.00 B 411x347 1690491015486660.png)
>>16152 however you did this, we need more
>>16195 Dude, FANTASTIC JOB!!! EXACTLY my type! I follow you on DA. How come you don’t post these magnificent images there?
>>16152 This is fucking amazing!!! Hope to see more like this!!!
>>16152 easily one of the best pictures ive ever fucking seen. keep the mega blobs coming dude
>>16152 >>16195 Great stuff here. Just followed you on DA, hope to see a lot more stuff like this. That immobile blob is impeccable, just pure sex.
(3.11 MB 1616x1100 HeavyBed3.png)
(3.42 MB 1646x1183 heavyCouch4.png)
>>16223 Honestly, I use these threads as proving grounds for my work. This is a very focused audience: a BBW/SSBBW board with image AI subthreads. So anything I post here will be towards literally the most receptive audience I could find for this kind of thing. Some of the stuff I post here isn't quite up to the level of quality I keep on DA. But I share pretty much everything I generate with you all since I know you'll enjoy it. >>16194 I feel like this is a ways off, but I look forward to the day. >>16070 It's morph work assisted by SD. >>16172 >>16202 >>16224 >>16228 >>16249 To the rest of you, thank you for the kind words, and I'm glad you like them. Megablobs are hard to get even remotely right, for every one I've posted here I've thrown away like 10. it's REAL easy for them to descend into flesh nightmares (and not the sexy kind). Full disclosure: HeavyCouch4 used this as a seed image: https://www.deviantart.com/ssbbwllover/art/Chizuru-Ichinose-Rent-a-girlfriend-wg2-972403696
>>16152 I mean, it does look like it could be the USS BBW.
>>16428 Truly a man of the people
>>13641 Really hoping to get this one on a realistic ai drawing
Is there any way to request a super sized morph of Jenni Bombshell BBW? She’s someone I’d love to see much larger!
(1.05 MB 1024x1024 SarahHuckabeeFatass.png)
(1.31 MB 1024x1024 00017-851727595.png)
Keep posting those fat-ass lard buckets! <3
>>16428 Can I get a catbox for this? Wanting to know what models and prompt we're used!
>>16104 Dying to know what models and promts we're used here too! Got a catbox?
(2.40 MB 1610x1002 HeavyBed4.png)
(2.63 MB 1541x1026 HeavyCouch6.png)
(2.32 MB 1698x1131 HeavyCouch5.png)
>>16831 >>16832 I think you've popped up today in a couple other threads I've posted in. Unfortunately, catboxing these images won't do you any good, as all my work is composite in nature. I've written a guide on how I do all this (link below), but here's the gist of my workflow: Generate an image at a low resolution. Edit that image to get closer to what you want to see, then reprocess it in img2img multiple times, slowly increasing the resolution. Edit the resulting images to help guide the process, adding and removing LoRAs along the way. Touch up the final image by reprocessing specific parts like hands and faces, then edit the final result like you would a photograph. https://www.deviantart.com/theguywhodidathing/art/How-did-you-DO-that-989204336
Thank you, awesome work!
What AI software are y'all using? Whenever I try to generate obese AI girls it seems to want to slim them down and make them just chubby.
>>16986 base64 encode the prompts.
(1.36 MB 1173x782 JennyBombshell1.png)
(1.58 MB 1403x782 JennyBombshell2.png)
>>16986 On the off chance you didn't get it, >>16990 is just kidding. In this thread, you're going to find people are either using stuffer.ai, bing, or Stable Diffusion running locally. I personally use SD with A1111, and I linked a guide here >>16921 >>16799 Oh, another throwback! Good call.
>>17030 My bad, that is just a jailbreak for the online stuff.
(1.00 MB 1024x1024 00020-3038982877.png)
(5.98 KB 240x240 1700969323007421.jpg)
>>16921 these are otherworldly good, you're an absolute pro
Any chance you could work your magic and do something like this?
>>17220 If I said please, no make it pretty please, would that help? 🤣
(3.60 MB 1702x1139 HeavyBed5.png)
(147.64 KB 300x264 HeavyCouch7-thumb.png)
>>17220 >>17446 If you can tell us who the artist is and for bonus points link to their portfolio, then fine. I'll put my money where my mouth is and provide a thumbnail now, though. For everyone else, here's another blob.
>>17452 The artist is chubby_tubby_3000 on IG. He draws blueberry lasses, of which I’m not a fan. The pic I asked you to do, the large lovely lady was blue. Here are a few others I changed the tint on. Feel free to tackle one or all of them LOL! Seriously though, THANKS for what you did!!! That pic is absolutely AMAZING!!!
(2.93 MB 1502x1320 HeavyCouch7.png)
>>17462 Alright, fair play. I'm sensitive to stepping on artists toes, so I'm glad that people who might be into that can find them now.
(1.36 MB 2852x1405 IMG_7797.jpeg)
(1.88 MB 3791x1405 IMG_7798.jpeg)
(2.36 MB 4927x1405 IMG_7799.jpeg)
Hey TheGuyWhoDidAThing, do you think you’d give something like a weight gain sequence a try if that’s possible? This one is by Bamboo-Ale, would love to see if you could tackle it : )
(119.47 KB 671x588 stuffer.ai-1211367.jpg)
(240.32 KB 1024x680 stuffer.ai-1373271.jpg)
(187.45 KB 704x832 stuffer.ai-1348476.jpg)
(148.90 KB 704x812 stuffer.ai-1226539.jpg)
(154.02 KB 704x832 stuffer.ai-1200707.jpg)
(160.18 KB 768x512 stuffer.ai-1323435.jpg)
>>17468 Understood. Check out his page if you need inspiration. He draws some great poses, but like I said, as blueberries. If you’re not into that, play with the tint to change the color.
(2.41 MB 1242x1622 Boberry5.png)
(4.43 MB 1862x2462 Boberry4.png)
(1.77 MB 1907x1308 Strawbrryymilk-1.png)
There was a time when everyone was trying to make USSBBW Boberry images. Here's a couple I did, using some old morphs someone did years ago as the seed image. >>17474 Sequences are tough, as it's pretty hard to keep identifiable features consistent across multiple generations. It's why I tend to stick to before and afters (see attached). But I'm usually up for a challenge, so I'll see what I can do. I think I'll leave Bamboo-Ale's art alone and try to generate something whole cloth instead, though.
>>17825 Haha! I'm the guy who made the original Boberry morphs you used. These AI versions look fantastic! Well done, man.
>>17825 Can you do more inflation?
>>17468 Dear The Guy Who Did A Thing, I’ve been a good boy all year. For Christmas I would like you to work your magic on either one of these images! To prove I’ve been a good boy and I know you don’t like to step on artist’s toes, the individual that drew them is on IG & goes by chubby_tubby_3000! Your friend, The Guy That Loves The Guy Who Did A Thing’s Artwork
>>18147 I never said Please and Thank You!!!
(2.89 MB 1413x1350 HeavyTube1.png)
>>18147 >>18162 Merry Christmas, ya filthy animals.
>>18184 Absolutely fuckin’ BRILLIANT!!! Thanks!!!
(935.06 KB 1024x880 00023-3003855632.png)
>>18147 >>18184 And a Happy New Year!!!
>>18431 Ok but besides the krita (where) do u (use) Lora And then the free automatic1111to do pieces of cleanup running off of the base model biggerGirlsModel_modelBV2.ckpt.
>>18441 I suggest moving questions to the question thread: https://bbw-chan.link/bbwai/res/2561.html
>>18424 You’re the man! Happy New Year!!!
(1.19 MB 1024x1024 00024-1477105047.png)
Some stuff with phone..
>>18497 >TheGuyWhoDidAThing Just read your DA post about your techniques today. THANK YOU! I have been making so many stupid mistakes. My top mistakes that TG enlightened me about: I have not been staging my LoRAs in order of coarsest detail to finest, also, I continued to use the same LoRAs or fine-tuned models in the refining stage, and finally, I have been using anime derived models in the upscaling. (I'm posting from my other computer while I am working out how to fit two air-cooled 4090's into my main PC. Tip to others: go with liquid cooled systems.) I have been brute forcing everything with inpaint. My main focus is on textgen, though, and my own improvement on "long term memory" for story writing. I am looking to use the MLewd Mixtral AWQ model, which is why I need more than 24G VRAM. I have got some story universes that will cause you to break your dick, and that people will be expanding on for decades.
(3.02 MB 1227x1843 HeavyRoom1.png)
>>18608 Hey, I'm glad to hear it was helpful to you! I'd love to see what you've come up with. I'd also love to figure out how to run LLMs locally on AMD hardware; I've been messing around on Caveduck and while it works well, I'd love to have the same level of control I do with SD.
(128.87 KB 868x665 IMG_0178.jpeg)
To The Guy Who Did A Thing: Any chance you could use your magic powers and create this but make her barefoot? The artist is Sirdishes on DA, just so you know. Thanks for any consideration!
(3.29 MB 1613x1256 SirDishes1.png)
>>19055 Why not. Broadly speaking, I don't mind 'skinning over' previously generated images nearly as much as I do actual art pieces, so this was fine. That being said, I wish I had a foot fetish sometimes. I get a lot of requests that include feet and I just... it's fine, I'm just not a good judge of what makes a good looking foot, y'know?
>>19060 Absolutely incredible! You’re doing the Lord’s work my guy, thank you!!!
(2.19 MB 1116x1292 HeavyGroceries.png)
(2.37 MB 1326x1326 HeavyMall7.png)
(3.16 MB 1554x1306 HeavyPark.png)
>>19061 You're very kind. I figured this thread could use a little color as well.
>>19068 These are absolutely incredible!!! You’re the Picasso, Michaelangelo & Bob Ross of fat babes!!!
>>19068 These are fantastic!!
>>19068 hats down, truly impressive work, Sir
(140.69 KB 739x521 IMG_0272.jpeg)
Anybody willing to put their skills to the test and attempt this piece from Tashoelle?
(2.93 MB 1434x1434 HeavyField1.png)
>>19087 >>19124 >>19146 Thank you all kindly!
>>19418 Another masterpiece!!! Just curious, how do you choose the background? Like on this one, what made you put this lovely lass in a field, as opposed to a bedroom, kitchen or by a pool? And please don’t think I’m criticizing, I’m not by any means. Just genuinely curious.
>>19470 Its speaking. Are you listening 👀
>>19471 Fuck are you talking about Dickweed?
(2.42 MB 962x1476 dex7491b-1.png)
>>19418 Because the three options you listed are the three most common backgrounds for this kind of thing. Most porn and porn adjacent photos are often shot in places like that (or soundstages made to resemble them) and that gets stale from time to time. If I have the ability to generate literally* anything**, why not get more creative with it? That's why you'll see landscapes and mall interiors: I'm particularly drawn to them. Plus, more complicated compositions allow for more detail opportunities that help 'sell' the illusion. Like, in that image you'll notice the blades of grass that are pressed against the bottom curve of her gut: even though this image is not the most photorealistic thing I've made (look closely at the rest of the grass, it's a little painterly), fun details like that help push your brain to accept them as real. But to make a long and entirely boring story short, I try to create interesting compositions that happen to feature impossibly fat or ludicrously curvy women. * euphemistically ** anything that the models I utilize have context for
>>19493 Thanks for the answer. Like I said, just curious. And great pic with your answer!!!
(584.25 KB 234x170 1703611139828557.gif)
>>16152 TGWDAT, you're a complete legend. When are we gonna get more like this?
>>19703 cosign, those ultra blobs are so hot and I want so much more, even bigger even
(2.17 MB 1326x1326 MDiazTorres1205-7.png)
(2.20 MB 1418x1418 HeavyStudio3.png)
(3.03 MB 1536x1239 HeavyBed6.png)
>>19724 Very, very good, thank you. But could be even bigger, fatter, wider. And then bigger still :) I love these though
>>19724 Ignoring the horny comments for a sec, those are seriously impressive blends of shop and generated. I know there is no way it's pumping something that perfect and shapely own it's own so your work is still incredible.
(1.92 MB 1296x1620 CypriotMami1.png)
(3.29 MB 960x2049 CurvyElizaBBW1.png)
(1.21 MB 828x1460 Squishy-Brat1.png)
(2.12 MB 1440x1800 AimeeTurley3.png)
>>19748 Thanks for the kind words, and keen eye! But, to be honest: every pixel in the above images is 100% generated. I wrote a guide (which I've linked here: >>16921 ) that explains the process, but broadly speaking you're right. There's no magic combination of words/checkpoints/LoRAs that will get you images that look exactly like this. The metaphor I use in my guide is tending a bonsai tree: you gotta gently prune and guide the 'growth' of the image through a series of steps, and touch things up in photoshop (or image editor of your choice) as you go. I've attached a couple of SD enhanced morphs I've done just to provide a comparison. In fact, that's the entire reason I got into learning SD; I wanted to make my morphs look more realistic. That's honestly the real power-usage of SD in my opinion. It's a great tool for fine tuning photomanipulations and collage work. The 'whole-cloth' generations are just a fun by-product.
(170.90 KB 1024x680 IMG_5911.jpeg)
(186.29 KB 1024x680 IMG_5910.jpeg)
(187.10 KB 1024x680 IMG_5913.jpeg)
(180.35 KB 1024x680 IMG_5912.jpeg)
After a lot of trial and error, I think I’ve finally gotten stufferai to reliably make decent USSBBW/immobile stuff without it falling into the “standard” shapes that it seems to like. I’ve started trending into even larger than what I’ve shown here, but I’m not sure if it fits the board or not so I’ve held off Last image spoilered since it’s a bit rougher than the others but I still like it
>>19749 Jesus christ this is immaculate
>>19825 Glory for mall bimbo goes to TheGuyWhoDidAThing, I own nothing yet owe everything.
(6.77 MB 1920x2268 BimboMall7.png)
(2.15 MB 928x1388 HeavyRoom2.png)
>>19749 >>19825 >>19846 Here's a new one just for y'all. Plus I seem to remember someone asking for really fat arms?
(3.10 MB 1281x1620 HeavyOcean2.png)
(6.63 MB 1792x1967 HeavyOcean1.png)
And here's a couple seaside images.
>>19808 Hey you know what. For using a simpler setup with limited pass options that's not bad. I think 3/4 of those are pretty great and the spoiled one is more legible than some of the garbage I made trying to mess around with this. Though the uniboob is a funny concept for that first image and a bit weirder than I think it should be. >>19932 Ok I swear that first image looks so familiar. Tell me that's based on a drawing cause I swear it looks like a photo real of a drawing I can't put my finger on. Looks awesome as always though.
Made with stufferai Had to experiment a bit with the wording to get the sizes right and proportion. Still a little weird in places
Weight gain progression + bonus sex POV
>>19932 That first one is so fucking hot, imagine what she'd look like even bigger, fuck
>>19932 god that first one is so crazy hot, imagine if she were even bigger, fuck
(59.06 KB 832x1216 00199-2233911373.jpg)
(102.39 KB 832x1216 00015-2871896361.jpg)
(97.86 KB 832x1216 00022-3340034391.jpg)
>>19931 Hi, TheGuyWhoDidAThing! I came across a funny puzzle and would like to learn from you and at the same time share my experience. Have you ever thought about squeezing the fat girl between any objects without resorting to any other tools besides the prompt? For example, as if she doesn’t fit into the doorway and she has to squeeze through. The goal itself is to depict a flattened body sandwiched between something. The catch here is that, as far as I understand (but this is not exact), SD does not have enough parameters to recognize through the prompt what exactly needs to be done. I think I've tried every adjective related to the effect of force on the body. The result was to carefully adjust the size of doorways/corridors and other spaces to fit the size of the body, even slightly larger than the body. Sometimes hips slightly squeezing between objects, but nothing more. Next, I went in reverse and tried to force the body to press on objects around. I was able to do this systematically in one direction (when the girl presses herself against something), but not in both directions. Yes, in a large number of generations, randomness throws up the correct result, but I attribute this to randomness. Since I don’t speak English very well, it’s difficult for me to form interesting verbiage, maybe I just don’t know more precise tokens like “squeezing through, stuck between, squash, condensing, compressing” and so on. There is an assumption that for “strong squeezing” there is simply not enough data in the models, but it seems to me that there is a possibility with them to “find a common language”. Any ideas?
(1.47 MB 1024x1280 grid-0068.png)
(1.91 MB 1212x1500 HeavyDoor.png)
(4.70 MB 1760x2208 HeavyForest1.png)
>>19961 I didn't use a base image for this one, no. But I certainly hope this isn't a drawing that already exists! If you do happen to find it, let me know... >>20135 Glad you like it. >>20159 I can say the idea has occurred to me, but I have to admit I'm not very interested in it. What you're proposing sounds like a limited toolset art challenge, like "Create a painting using only four colors and the clone stamp tool." I'm sure someone would be able to do something cool with a challenge like that, but it's not for me. This is something I've talked about before regarding the way we approach StableDiffusion and other image 'generators': nearly everyone thinks of them like magic image machines where you type in the right words in just the right way and then they give you exactly what you want. I honestly believe that's not going to happen anytime soon, and it's definitely not how they work right now. You've touched on the primary reason why, though: there isn't enough data for something like that in most models. I'm not an AI expert, but my understanding of how this all works is it's effectively auto-complete for collage work. When you type something like "beautiful woman, large breasts, bikini, pool" SD trawls through a context database to find images (or rather the concept of images, but we're getting into the weeds here) that share most of those tags. after generating noise, it tries see patterns in the noise and 'draw' lines and colors that have similar shapes and values to the images it's referencing that connect the points it 'thinks' it 'sees'. It's not literally doing collage work (cutting parts of images out and gluing them together), it's a bit more like tracing. But at the end of the day, SD can only 'trace' things it's 'seen' before. That's why if you use the example prompt above, you'll get a pretty nice looking image even at a low resolution: the core SD model has seen billions of images of conventionally attractive women in bikinis with 'large' breasts, so it's really good at spitting similar images out. If you type "A 1000 pound woman squeezes through a doorway, gigantic belly, doorway, stuck" it's going to loose it's mind trying to make sense of those words. That's also why there are so many specific checkpoints that have been created for use with SD. People are basically creating new context sets for SD to utilize. To further stretch my collage metaphor, say you wanted to create the image you're describing by cutting out pictures from magazines and gluing them together. If I hand you a stack of Maxims and Playboys, you're going to have a pretty hard time making it work. But if I gave you the full print run of BUF and other BBW focused magazines, you'd be much closer to your goal. This is why I utilize the img2img functionality of SD. If I can take an image I already have (whether it be a real photo or a rough image generated with SD) and do some fairly simple edits to it, I can feed that back into SD and tell it to keep that shape in mind. The following generations already have a base image to reference, and it can recognize elements of the image to work from. So, until we live in a world where the average woman weighs 800 pounds and everyone gets turned on by seeing them struggle to fit through a doorway, we're just not going to have enough resources to pull from for SD to be able to create a similar image accurately from only a prompt. Even when you include art (while not a massive dataset, still enough to get the general idea of a massively fat woman across) what you're looking for is pretty niche, and I would (generously) guess there's only a couple hundred drawn images to draw from, AND that's assuming they've all been compiled into a LoRA or a checkpoint somewhere. So, the image I've attached here is an example of what I was able to do in the past. Using (a now old version of) Symix, the creator of that LoRA was able to tag some images with "stuck_in_doorway" which I call in the prompt: >a fatblob woman is stuck in a doorway, stuck, stuck_in_doorway, fatblob, gigantic belly, huge breasts, fat face, two tone lighting, dimly lit, low key, <lora:epiNoiseoffset_v2-pynoise:2> <lora:symix-preview-4-9:1> Out of this grid of four images, only two of them have the 'pressed against the doorframe' look, but none of them have that 'squish' you're looking for. the next image is the final result after editing the third image in photoshop and reprocessing it multiple times. I had to edit the belly to make it larger, and curve the sides to give the impression of her fat spilling out 'through' the doorframe. That's also how I would approach your challenge. Rather than try to get the prompt to give me exactly what I want, I get something that's in the vicinity of the image I want to make, then edit it and feed it back into SD. Repeat this process until you have an image you're happy with. So, I hope that helps somehow. I've included a link to the Symix LoRA that includes a "stuck in doorway" tag, but understand that it won't just magically work. I'm just handing you a magazine that has a few images of fat women stuck in doorways, it's up to you to make it work. And finally, what checkpoint/LoRAs did you use to generate the images you've shared? I don't recognize the style/shapes, and I'm curious what you did to get them. To everyone else, a new random image for you.
>>20206 shit, forgot the link https://civitai.com/models/121416
TGWDAT can you make more fat goth blobs please?
>>20206 That's why I'm a bit sad over how gimped and censored DALL-E3 has gotten over the past several months though- it was probably as close as you could get with AI image generators in terms of "type in the right words and they give you exactly what you want." At least compared to SD, it had a much better overall understanding of taking in everything you were putting in a prompt and even generating bigger sizes, all without the need for extra LORAs or custom models and the tinkering you need to get a nicer image in SD. But like you said, I think img2img with something rough generated in Dall-E3 and then tweaked in SD is about as easy a workflow you can get right now
(1.42 MB 1024x1024 00043-780867075.png)
(2.09 MB 1280x1280 00063-1262204398.png)
(1.78 MB 1280x1280 00104-490929298.png)
>>20206 Thank you for such detailed thoughts, I am aware of how SD works under the hood. Just regarding this example, I wrote that “this is not exact” because the model is 100% trained on “images with stuck girls”, it’s just that the number of parameters for “detailing” the composition is not enough due to the limitations in interpreting prompts for translation into numbers (unlike from Dall-E 3, which has its own interpreter trained on spatial examples, which is why he is so good at arranging objects in the space of a composition). SD developers increased the number of parameters in the XL model (6.6 billion compared to 0.98 billion for the 1.5 models), which partially improved the situation (especially the hands and legs), but... Yes, SD still works like “I draw what I remember”. Therefore with the help of prompt engineering it is possible to direct “diffusion itself” in the direction we need a little more than simple tagging (for example, to separate the colors of different objects). We can add additional tags to the prompt to simulate that, what we want, bypassing the direct description of the phenomenon (for example, “motion blur” instead of “high speed of movement”). You can endlessly try to make a pot-bellied girl with skinny legs, having “skinny legs” in a positive prompt. But you can simply add "lipedema" (a disease of fat accumulation in the lower limbs) to the negative prompt. Such tricks with prompts will allow you to work less with “post-processing”, you know what I mean? I didn't mean this as a "Create a painting using only four colors and the clone stamp tool" challenge at all :) I simply emphasized that I mean only the result from the prompt. I am aware of the possibilities of inpainting and other methods, Yes, for high-quality results now there is no way without post-processing, but I just perceive it as one side of a two-step process (automatic generation and manual post-processing). The quality of the first depends on the volume of work of the second. Therefore, I consider prompt engineering skills to be an important part of the process too, as it can greatly facilitate manual work. And so I was interested to ask you about this, maybe you could suggest something. Oh yes, thanks for Symix LoRA, by the way, I've been using it since version 3-1, when it wasn't on Civitai yet :) And yes, you asked about the checkpoint\LoRA I use. Now I am actively mastering the capabilities of SDXL and until recently I was waiting for checkpoints/LoRA trained on fatties/blobs to appear. And now, finally, one full-fledged option has appeared. The example images I presented in this post and the post above were made on SDXL only using a prompt without LoRA and manual post-processing (this is noticeable about post-processing, I think, hehe). I wrote about this here: https://bbw-chan.link/bbwai/res/17618.html#q20025 If your PC specs allow you to run XL, then I highly recommend giving it a try. Prompt adherence surprised me very pleasantly.
(2.84 MB 1649x1098 FullField3.png)
(3.00 MB 1264x1488 FullQuarry1.png)
(3.39 MB 1664x1395 FullBed5.png)
>>20349 Ah, gotcha! Yes, you're right about Dall-E 3, it's the only one that even comes close to the idea. I still stand by my statement that it can only go so far, but my post was already way too long for me to go into it. Sounds like we're generally in agreement. And while I certainly understand what you mean about prompt reliance, I still enjoy the level of control I have over the image generation process with my approach. Prompt engineering is definitely a skill that can be honed, but I prefer to work from a visual basis, rather than a textual one. To each their own! And unfortunately I can't get SDXL to run correctly on my system. I'm on a Radeon VII, so I'm already gimped with an AMD chipset, and it only (ha, *only*) has 16 GB of VRAM. But good luck to you! I hope you are able to figure this one out! For the rest of you, I don't have any new SSBBW images at the moment, so here's some huge tits in penance. >>20255 No promises but I'll see what I can do.
(2.25 MB 1554x1248 HeavyRoom3.png)
(3.61 MB 1555x1555 FullStreet1.png)
>>19734 >>19748 >>20255 Not a goth or a bimbo, but definitely a blob. Plus a couple random absurdities.
(2.08 MB 1500x1500 05501-1880351695 final.png)
(2.80 MB 1722x1722 05495-2388147258 final.png)
(2.22 MB 1500x1500 05526-3241416818 final.png)
(2.29 MB 1500x1500 05523-3156579145 final.png)
>>20512 Followed some tips on that guide you posted and I'm pleased to report that The Method works pretty well. Still not 100% there yet but here's a few I put together today
>>20512 oh my god that first one is the hottest stuff ever
omfg where do u find these bitches? thank god for america lmao but serisly, theseare like my old fantasees id have of giving girls at school endless cake and burgers and shit until they about to puke, then give them pepto then more food and in my dreams i also put more cals in there IVs goin straight to bloodstreem
>>20512 >>20543 I've also gave it a go and looks really cool your method TheGuy! Impressive find! Just one question, with every hi-scale step (e.g. 1.2x) it also blurs the image more, so after 3-4 steps is very blurry. Have you experienced that too and you know maybe how to avoid that?
(3.40 MB 1421x1421 HeavyPool1.png)
(1.58 MB 1152x864 Sadie3.png)
>>20650 Honestly, it just depends on what your workflow is. For me, in Photoshop, I'll upscale by 1.2 using Preserve Details 2.0, and then sometimes I'll select the subject and add another layer of sharpening if I feel it needs it. If you're only using SD, try reprocessing the image at a fairly low denoising strength, like 0.2-0.3. you can even use the 'loopback' script to automatically loop this process a couple of times (it runs img2img, takes the resulting image, and feeds it back into img2img for the number of loops you define). That can sometimes bring more clarity to an image. For realism, I've found that sometimes adding 'natural lighting' or 'studio lighting' can sometimes help SD focus a bit on details (depending on the nature of your image). It's hit or miss, but worth trying as well. (images unrelated, second one is a morph with SD assistance)
(3.50 MB 1638x1638 HeavyBar1.png)
>>20255 One Big-titty-big-belly-three-chair-ass-having Goth GF out for sushi.
(24.00 KB 679x484 1697304108824875.jpg)
>>20512 >>20669 good grief man, these are so absurdly good. please give us the fattest you can make
>>20659 Thank you very much! I think I'm finally getting the idea plus once again carefully double-checking with your tutorial. The blurry effect was both due to using low denoising value (surprisingly) at ~0.3 and mixing some of the anime + realistic loras with using a realistic checkpoint. Leaving only realistic loras and increasing denoising to ~0.4 / 0.45 helped here to improve the realism of the image. Or, when using denoising at ~0.25, to sharpen existing details. For the realism and lightning keyword - good point. I also add sometimes LowRa lora which plays well with certain lightning keywords. Still, hats down to your showcase, your gens are a masterpiece!
(1.11 MB 888x888 Absurd2.png)
(3.88 MB 1766x1766 Absurd3.png)
>>20134 >>20135 >>19734 I swear, some of you gremlins don't have a limit.
>>20732 YESSS... YESSSS!!!! Never enough fat
>>20699 You've definitely got the hang of it. Hope to see more of yours.
>>20764 Yeah I got it. Ill do it myself.
>>20699 DAMN!!! Talk about quality!!! Hope to see more of your stuff in the near future! And hopefully some full body stuff! This is exceptional!
(13.01 KB 647x607 1689639993950099.jpg)
>>20732 absolutely cannot overstate how amazing that first one is, im in utter shock i would pay you to make more like that, and even fatter
>>20802 the face in the second one is maybe just the tiniest bit scuffed but thats the one thats really really making me go nuts, insanely hot
>>20732 I didn't think it was possible to nut as hard as I did at that first picture, I nearly passed out lol MORE BIGGER FATTER
>>20732 11/10 Supreme work from you again. Really would love to see more of these hyperfats and megablobs as realiztic as possible
>>20732 holy shit plz go bigger!
(157.32 KB 1024x680 IMG_6001.jpeg)
(145.50 KB 1024x680 IMG_6002.jpeg)
(135.02 KB 1024x680 IMG_6003.jpeg)
(176.19 KB 1024x680 IMG_6004.jpeg)
(182.12 KB 1024x680 IMG_6005.jpeg)
Since you guys seem to like it given the recent responses here, have some hyperblobs that I managed to make, although strangely it seems that the model likes to delete breasts, and when I force it to not do that it gets rid of the arms lol
>>20861 I should probably learn to refine these with the TGWDAT pipeline at some point, but that’ll have to wait for some other time
>>20861 ....okay, now inflate them....some....fill out them creases,
>>20861 In some way I kind of like that you can’t see the breasts on most of these, like they’re so big that they’ve just been swallowed up by fat… also some girls when they get to a certain weight their breasts sag in a way that the nipples end up being hidden by the their belly anyway so it sort of works in reality!
(3.29 MB 1480x1525 HeavyRoom4.png)
(2.54 MB 1320x1320 Absurd4.png)
>>20618 >>20672 >>20734 >>20802 >>20805 >>20832 >>20833 >>20849 Alright you fuckin' gremlins, here's some more insanity.
(2.23 MB 1026x1543 HeavyPool2.png)
(2.39 MB 1101x1658 WidePool3.png)
(3.42 MB 1220x1841 WideStore1.png)
>>20699 Thank you for the kind words, and yes, I'd say you're getting the hang of it! Looking forward to seeing what you can do. >>20543 I'm glad it helped! You're already achieving the same level of quality I was hitting a few months ago, so you're well on your way already.
>>20732 More? Could you do a redhead girl who's just absolutely an ocean of freckled blubber?
>>20986 I think a view from the cieling towards the girl would be amazing and also with bigger faces ...even better
>>20963 Sir, these underwater gens are unbelievable, every time you raise the bar even higher, wow. Being able to generate both underwater with proper lightning and skin pores, all the details, it's just mind blowing. Loopback Scaler works pretty neat so far.
(2.91 MB 1227x1546 HeavyOffice5.png)
(3.11 MB 1548x1032 Absurd5.png)
(3.42 MB 2072x1376 Absurd6.png)
>>21044 You're very kind. Here's the secret: https://civitai.com/models/29491/jk-underwater-naked-shot Mind you, it *really* doesn't want to generate faces underwater, so you have to kinda work at it. There's a few other LoRAs like it that I haven't tried yet, so definitely experiment. And your details are coming along! don't rely too heavily on the loopback script though, as it does have a tendency to start smoothing out skin details. It's all a balancing act.
Anyone thought about experimenting and generating some SSBBW/USSBBW’s with specifically hyper fat areas? Maybe some more unique parts like fupa, back fat, arms and legs etc… might lead to some interesting looking blobs : )
>>21114 This is so hot. Is there a way to crank up the freckles to absolutely insane levels? Like "Autumn" by Tonbelly levels of body coverage
(2.48 MB 1380x1380 FeckleTest.png)
(4.24 MB 1468x2062 HeavyRoom5.png)
>>21230 hmm... I think the best answer I can give you is maybe. Attached is a quick test I ran to see how far I could max out on freckles without painstakingly painting a million freckles by hand. You can see the result is pretty good in my opinion, but it's real easy for it to slip into skin cancer territory. However, when generating absurd blob sizes, SD barely recognizes the shapes as skin and I have to fight to keep it from interpreting random folds as cushions or fabric of some kind. keeping minute details on the skin is quite difficult. But it's interesting to experiment with, so we'll see if something comes of it. >>21228 I encourage others to try. Hyperblobs aren't even my 'thing' in the first place, I just like trying it out sometimes. Getting into asymmetric bodyshapes like that is too far for my tastes, unfortunately. Second image just for variety.
>>21114 we need more of this "absurd" 5 and 6 ...and probably in more situations or environments
(3.61 MB 1488x1488 HeavyMall7.png)
(3.40 MB 1226x1543 HeavyParking4.png)
(1.91 MB 1008x1158 FullBakery1-s.png)
>>21233 I wouldn't hold your breath, I'm a bit burned out on hyperblobs at the moment. But when I try again, you all will be the first to know.
>>19750 Love the Aimee turley one
>>18184 Veins are brilliant
(203.94 KB 1024x1024 ComfyUI_temp_mynbi_00164_.jpg)
>>21376 >I'm a bit burned out on hyperblobs at the moment. I'd be honored to step in.
>>21421 Don’t stop there! Please continue!!!
>>19493 TGWDAT, if you’re interested in generating specific ideas and exploring different backgrounds, I definitely have a request based on some smut the lady friend in my life shockingly came out with the other day that I’d love to see. If you’re still taking people’s requests, the scene is two huge hucows in a barn or barnyard setting, spilling out of cow-print bikinis, cow ears, etc. One of them a standard-blob sized blonde with an emphasis on her tits, and the other a curly red-headed ~600lb USSBBW with huge thighs and belly!
(167.21 KB 1024x1024 00026-2048584176.jpg)
>>21425 Thanks! :-)
>>21585 Are you able to do more of this please, this is god tier stuff
>>21588 cool that you liked it. Another try, still learning and testing the overall process and limits of SD 1.5
>>21650 Long blonde haired blob girls is my weakness lol
>>21650 One of the best I've ever seen.
>>21650 This is simply magnificent!!! Please continue!
(792.66 KB 3072x3072 00012.jpg)
Some blob action for this sweet thread!
>>21585 >>21650 I have already learned so much from your two posts, and I have been working at it for six months! And you say you are new?!? Good god, man! Keep posting! You are an artist! I have also recently visited the BiggerGirl's discord, and have learned much by getting all of the tags and such. Also found the justfat lora there. I have some questions... What is the merge-3ni-bgmbv2-r04.fp16.fp16 model, and how do I find/create it? Also, same question for merge-bgmreal-stoon-buxom-3r035035.fp16. What is the difconsistencyRAWVAE_v10LOWSafestensor.safetensors VAE? Google hasn't been helping on these.
>>21751 Thanks for the kind words, really. And thanks for heads up with the BiggerGirls discord - it's been a while since I poked around there. I've been also playing with SD for a while, though I have a potato apple mac w/o proper GPU and doing it in very limited free time for fun, so cannot really test it properly. Since generating images takes too much time, I am careful with trying to find some good working prompt patterns and cherry-pick loras with custom merges. Then, I pick couple of seeds. As you may have seen, I use many loras with low strength, each pretty much cherry-picked to get certain styles / patterns / concepts. So that the final image is not over-burned but still certain concepts are present (e.g. detailed skin, fat rolls, cellulite, etc). The motivation behind custom merges is that many anime-style loras do not play very well with realistic checkpoints, same often applies for the opposite, for realistic loras. Moreover, some realistic loras are trained on different base models and so may not play well with other realistic models. So with the custom merges I am trying to find some sweet spot. Once again, many trial-errors with hand-picked ratios, as I do not have GPU power to run proper comparison in batches with different block weights, nor trying more advanced techniques such as bayesian merging. Then, after generating the initial image, thanks to TheGuyWhoDidAThing tips, I apply upscaling iteratively, with once again lowering down the strength of anime loras and increasing the realistic ones. What I am doing and suggesting here may be also a complete nonsense, I may be overcomplicating, due to lack of knowledge, skill, etc, but I just do it for fun and somehow it gives nice results. As for VAE, this one works pretty cool: https://civitai.com/models/88390/difconsistency-raw-vae-pack Regarding custom merges - here are some: https://huggingface.co/randon-mumber/fat-merges Have fun!
(5.07 MB 5000x3034 IMG_0082.png)
Anyone think they could try generating a shape like this? Obese lower half (doesn’t need to be a blob) but with a skinnier upper body, arms and face etc
(71.50 KB 1125x682 00017.jpg)
>>21871 Here is one done off of the image. (this approach is not recommended for blobs. Getting all of these elements right: hands, toes, face, the mini-fridge, etc. is like herding cats.) It is way easier to do from scratch. I will do more when I get back in town. The hyper bottom heavy lora is great for this.
(104.24 KB 1024x1024 00178-1589049821.jpg)
>>21871 Still learning...
(123.39 KB 1024x1024 00039-3665483837.jpg)
>>21871 Not 100% realistic, but still learning how to get blobs and realism to play nice together. Also, posting jpegs to not soak up all of our benefactor's bandwidth.
(98.67 KB 1024x688 00040-3665483843.jpg)
Getting back on topic, lol.
(89.79 KB 1024x688 00042-2949862895.jpg)
(97.86 KB 1024x688 00045-1249807702.jpg)
(114.23 KB 832x704 00047-2416571132.jpg)
>>22027 >>22034 May i ask for program and prompt?
(90.02 KB 832x704 00048-140771858.jpg)
>>22121 Sure! Automatic1111, and I installed the add-on Regional Prompter (easily done in the UI). I created two regions (50% top and 50% bottom) On the first image, I checked the box for common prompt, then I used the following prompt: A nude woman with a thin upper body and an enormous ass sitting on bed BREAK 1girl, thin face, thin arms, thin shoulders, huge flabby breasts, flabby breasts, breasts BREAK 1 girl, enormous ass, huge ass, enormous thighs, enormous legs, blob, hyper ass, hyper breasts, thick legs, deep navel I started out with the bigger girls model BV2, and may have had the hyperbottomheavy LoRA in the prompt to start with, both available on civitai.com. (Lost the full series of prompts used, but you'll easily get the hang of it with some practice.) I then upscaled it in the img2img tab after switching to the bigger girls REAL model and switching to the VAE mentioned here: >>21807 which contains other great advice along with the poster TGWDAT who has great advice in this thread, as well as the pinned tutorial thread where he links to an article he posted on DeviantArt. The keywords that the BG models were trained with can be found on the bigger girl's creator's discord. (link can be found on the civitai page for his models. You'll have to hunt for the keywords list and not ask for them or else risk annoying the creator who has performed several absolute miracles that very few people would be capable of. Shouldn't take more than 5 minutes.) On the second one, I used the justfat lora, which I learned about on the BG discord. It has three keywords, one of which is fatlegs. (You can often review the keywords associated with a Lora in A1111) The justfat LoRA is pretty random, though, so you'll have to generate dozens of images at a time to find a usable one, an NVIDIA card with enough VRAM is recommended. On the second one, I couldn't switch to BG REAL for the upscaling, and had to use model BV2. Also, I had to do some inpainting to remove a couple of weird artifacts that justfat gave. Everything else is going to have to be researched by yourself, it is just too much to cover. (e.g. the art of selecting the right CFG values, denoising strengths, etc. throughout each stage of the process.) Don't be disappointed with your results in the beginning. Keep learning! Also, the tutorials out there cover normal body types. Similar results cannot be expected for us.
(133.29 KB 832x704 00049-1136646682.jpg)
(121.04 KB 832x704 00050-2836735413.jpg)
(85.93 KB 1152x768 00071-1911064926.jpg)
>>22057 >>22082 The girl in the first of each of these photos used to suck her thumb until she became too fat to do so. Then she resorted to tucking her right hand under her right breast.
(68.58 KB 768x512 DamnFineAss.jpg)
I spent yesterday going through 143 keywords in bigger girls model BV2 and documenting the solo keywords results. I started with the keywords with the highest frequency and stopped testing when I got to keywords with 22 images or less. I skipped a lot of uninteresting keywords. I would generate 25 images with a CFG of 13. The results don't always match up with the keywords. For example: "ssbbw" is usually larger than "ussbbw", but the best one of all was "fat rolls". Also, I found it is way better to use the cellulite LoRA than to use the "cellulite" keyword, which makes the skin really bad. Some of the keywords, like "tube feeding" appear to be trained on a mostly loli images. I had a laugh when I put the embedding verybadimagenegative_v1.3 as the solo positive prompt. "hyper" results in giantesses, and if you ask for "hyper ass" and don't want a giantess, you may want to include the solo "hyper" in the negative prompt. Learned so much, I will share the list with descriptions once I clean it up (I was feeling very snarky when I was writing the descriptions yesterday.)
(100.48 KB 768x512 PigGirl001.jpg)
(102.29 KB 768x512 PigGirl002.jpg)
(96.59 KB 768x512 PigGirl003.jpg)
>>12302 In the year 2071, the dehumanization proclamation was announced. The near-human pig-hybrids were officially pronounced to be the property of their owners. They then became the property of choice of feeders. World records were then broken...
any chance you can try more taller ussbbw or looming angles?
(717.28 KB 784x856 00024-3845303114.png)
(700.61 KB 784x856 00027-314264967.png)
(722.82 KB 784x856 00032-4121588576.png)
(728.63 KB 784x856 00031-3420092014.png)
(708.49 KB 784x856 00035-140974254.png)
(722.22 KB 784x856 00036-58784979.png)
(666.68 KB 856x784 00042-1157329019.png)
(922.56 KB 856x784 00059-4128881065.png)
If anybody ever feels like taking any requests, what about some AI upgrades of BBW’s like Chloe, Aliss Boynton, Mochii into USSBBW’s, if it’s possible? Just throwing it out there, no worries if not doable.
PEARfectionist on DA
(105.84 KB 828x828 IMG_8048.jpeg)
(926.63 KB 2024x1336 00003-00002-3043809441.jpeg)
(721.05 KB 2024x1336 00006-00002-3043809441.jpeg)
(2.16 MB 1440x1440 00125-1727024285.png)
(2.28 MB 1536x1536 00172-2264591691.png)
(3.17 MB 1776x1776 00035-00035-00004-3800337739.png)
>>23669 Ayy that me. Could’ve at least put it in the bottom heavy thread xD https://www.deviantart.com/pearfectionist/gallery
(1.38 MB 680x1120 HeavyStudio4 (Large).png)
(2.94 MB 1400x1400 HeavyMirror1 (Custom).png)
(3.07 MB 1080x1634 HeavyCosplay1 (Large).png)
>>22432 Is it possible to do more long blonde haired girls please?
(1.61 MB 1048x1048 00000-3489396232.png)
South Korea's best female racer can barely fit into her jumpsuit these days, let alone a racecar, but she promises if you win tomorrow's race she'll try to get back in the driver's seat... while you're still in it.
(1.28 MB 1024x1024 blonde.png)
>>24484 what about?
>>24489 Thanks, i feel like theyre very underrated, not enough of blonde ussbbws imo
(1.44 MB 1024x1024 blonde-2.png)
(2.09 MB 1024x1536 blonde-4.png)
(2.11 MB 1024x1536 blonde-3.png)
(2.11 MB 1024x1536 blonde-1.png)
>>24499 posting then more blondies for you, as I am testing the new swimsuit collection for the upcoming summer
>>24554 Thanks so much, just a itty bitty request, could you do them with a bigger boberry style body please?
(1.25 KB 56x125 1705266469081948.jpg)
>>24471 that mirror pic is wildly great, tgwdat you have the craft down. in desperate need of more of the immobile blobs and fat faces from you
(1.82 MB 1024x1360 swim-1.png)
(1.77 MB 1024x1360 swim-2.png)
(1.78 MB 1024x1360 swim-3.png)
(1.64 MB 1024x1360 swim-4.png)
(1.26 MB 1024x1024 swim-5.png)
>>24562 some of them are kinda boberry, but some are rather more boobery, hope you like them
>>24913 God damn these are amazing, please do more :O
(2.07 MB 1024x1536 bbw-1.png)
(1.98 MB 1024x1536 bbw-2.png)
(1.08 MB 768x1024 bbw-3.png)
(1.84 MB 1024x1536 bbw-4.png)
(1.63 MB 1024x1360 bbw-5.png)
(102.24 KB 744x710 IMG_7823.jpeg)
Is there ANY chance you could recreate this, but instead of her standing, have her on her belly ready to be rolled? Also, clothes are NOT necessary. Hey, you don’t get if you don’t ask. Love your work and follow on DA as well. Thanks!
Big booty ussbbw
Please feel free to do something with this, can't quite seem to get it any better
>>25569 this is pure kino, more drowning blobs of blubber PLEASE
>>25752 >>25569 agreed, how was this done
I was just messing about with PixAI, I've got a few different pictures now, but the quality is a bit rough. Would love it if someone fine tuned them?
(1.26 MB 1280x832 ultra-pony-1.png)
(1.12 MB 1152x896 ultra-pony-2.png)
(1.16 MB 1280x832 ultra-pony-3.png)
(1.13 MB 1280x832 ultra-pony-4.png)
(1.12 MB 1152x896 ultra-pony-5.png)
Although not perfect at all, but after playing few days with PonyXL and without any highres fix, inpainting, etc, it's really hard to think about coming back to SD1.5
>>25971 Those are amazing! Do you have a DA account or any place where there are more?
(1.84 MB 1536x1024 ponies-1.png)
(1.83 MB 1536x1024 ponies-2.png)
(1.84 MB 1536x1024 ponies-3.png)
>>25973 thx! I don't have (yet?) any DA as this would put pressure to keep it updated and care more about stuff being publish there. I mostly run gens on my local potato laptop, edit here and there for fun and in spare time. Such as these ones.
>>26054 Please, put more of squashing, facesitting and riding stuff ! *.*
(1.05 MB 896x1152 bathroom-4.png)
(1.08 MB 896x1152 bathroom-2.png)
(1.06 MB 1152x896 bathroom-3.png)
(1.56 MB 1472x1088 bathroom-1.png)
in the meantime, a photo session from the bathroom
(727.64 KB 768x768 00050-4139462781.png)
(1.21 MB 1024x1024 00070-553432809.png)
(1.31 MB 1024x1024 00063-3831594966.png)
(1.32 MB 1024x1024 00093-277853177.png)
(1.38 MB 1024x1024 00084-3034483866.png)
(1.46 MB 1024x1024 00070-1939306488.png)
The amount of amazing naked blob girls in this thread is amazing. I'm trying this myself for the first time, does anyone have any prompt tips for getting a straight-on shot? Like... full frontal, not 3/4 turned?
>>26054 for the love of god, PLEASE make a DA. we need to archive these gems.
(1.62 MB 1472x1088 ass-1.png)
(1.71 MB 1472x1088 ass-2.png)
(1.58 MB 1472x1088 ass-3.png)
(1.70 MB 1472x1088 ass-4.png)
>>26519 will try, just don't know when, as I'm constantly blown away by the capabilities of this tech, lol
(1.28 MB 1024x1024 00021-3479025054.png)
(1.28 MB 1024x1024 00032-1328666354.png)
(1.52 MB 1024x1024 00016-3619007968.png)
(1.74 MB 1024x1024 00013-636457241.png)
>>26562 really nice cellulite and fat rolls! a small Q - if you increase CFG rate would the body shapes or poses become more natural?
(697.19 KB 768x768 00046-2941719777.png)
(790.69 KB 768x768 00039-1478968161.png)
(769.25 KB 768x768 00065-1373352906.png)
(1.30 MB 1024x1024 00032-00362-3508241459.png)
(1.29 MB 1024x1024 00066-2185633337.png)
(1.42 MB 1024x1024 00040-00370-3508241467.png)
>>26563 i left my computer generating thousands of these overnight with randomised parameters and tags to try to crack the obesity code (CFG also randomised). there were good ones but there were also a lot of disfigured blob monsters. promising results but more research required. here are some more, and a failure example.
(1.14 MB 1152x896 fat-1.png)
(1.12 MB 1152x896 fat-2.png)
(1.13 MB 896x1152 fat-3.png)
(1.74 MB 1088x1472 fat-4.png)
>>26567 Which one is the "failure"? They all look good to me.
Got to try generating some where they’re tall enough to barely not have their heads bump the ceiling
(1.11 MB 1152x896 big-1.png)
(1.15 MB 1024x1024 big-2.png)
(1.06 MB 1152x896 big-4.png)
(1018.52 KB 1152x896 big-5.png)
(1.70 MB 1472x1088 big-3.png)
(165.60 KB 704x832 IMG_9755.jpeg)
(1.21 MB 896x1152 swimsuit-1.png)
(1.86 MB 1088x1472 swimsuit-2.png)
(1.38 MB 1024x1024 swimsuit-3.png)
(2.10 MB 1472x1088 swimsuit-4.png)
(1.37 MB 1152x896 swimsuit-5.png)
(2.04 MB 1472x1088 swimsuit-6.png)
a new swimsuit collection has just arrived
>>14394 >>16428 What program did you use for this?
>>14394 >>16152 If I gave you an image could you make them as big as this?
Any chance someone could create more super realistic USSBBW pics? Some of those here are absolutely hot.
I'd love to see some more ultra fat beautiful girls just laying around and filling up the whole bed, barely able to move.
Can anyone make a hyper pregnant ussbbw?
(1.93 MB 1152x1152 HeavyStudio6.png)
(2.22 MB 1384x1336 WideDoorway1.png)
(2.27 MB 990x1287 HeavyStudio5.png)
(881.13 KB 1728x1656 BeachedWhale.jpg)
>>28641 whenever TheGuyWhoDidAThing replies for any of the asks it's a pure gold. impressive stuff!
>>28605 Bumping for anon having taste
(1.15 MB 1024x1024 00043-893020177.png)
(316.50 KB 1240x1856 WideLivingroom1.jpg)
(920.14 KB 1952x1468 HeavyOffice6.jpg)
(2.78 MB 1478x1447 HeavyBed8.png)
>>28661 you're too kind.
(293.42 KB 1406x1075 HeavyLivingroom1.jpg)
(195.62 KB 1280x1280 HeavyBed8.jpg)
>>25781 only like three months late, but it's something... >>12302 it's also been like a year, figured I'd take a shot at uprezzing one of the original images that started things off.
>>28678 >>28717 dear god man i can only get so hard. these are pure heat, your blobs are divine. would love to see this girl as a realistic blubber blob
(349.07 KB 1608x1189 HyperBlob4.jpg)
(323.98 KB 1536x1152 HyperBlob5.jpg)
>>28723 fuck it, why not.
>>28745 that is unironically the best blob i have ever seen, i think im seeing stars, the head and blubbery body came out absolutely perfect. this is beyond top tier, keep up the heroic work
+1 - definitely a talent for bringing those enormous fantasy-sized fatties to life! I haven't really seen anyone else successfully get AI to cope with that kind of scale/detail. I'd love to see more - there are plenty of iconic ultrasize scenes that could be incredible with a refresh. Great work anyways.
>>28745 I LOVE your hyper blobs SO much. By the way the old deviantart is gone, do you have another?
(328.94 KB 1500x1000 HeavyRoom6.jpg)
(478.56 KB 1584x1598 HeavyBed9.jpg)
>>28747 I'm glad you like it. For everyone else, make sure to go check out https://www.deviantart.com/hypnagogum as his work is quite nice. >>28767 Thank you kindly. Feel free to share any other references you want me to take a crack at. Just make sure to cite your sources where possible. Funny thing is that hyperblobs just aren't my thing at all. I like my women big, but these sizes are *a lot.* Turns the generation process into more of a puzzle for me to solve than anything, which is what I enjoy in the first place. No shame/shade to those of you that do, of course! Everything is somebody's something, and that's pretty neat. >>28785 Yep, unfortunately it's permanently removed. Got a DMCA strike and then a manual review led them to believe my work was improperly obtained and they shut it down. Maybe I'll move to tumblr or X or Bluesky or something. But for now, I'm just here. Might upload my entire catalogue as a zip somewhere sometime for anyone who wants it.
>>28821 That's a bummer with DA, they are a mess. Please give a chance to tumblr, we'd love to see more!
>>28821 If you ever do make a twitter or whatever please post it here, I'm probably your biggest fan, I could feed a couple of these blobs with how much rope I'm blasting
Big bbw girl who loves a skinny man and love to dominate him
(73.86 KB 768x512 BBWChanTryHarder.jpg)
(76.79 KB 768x512 BBWChanTryHarder2.jpg)
(723.73 KB 1499x2042 WideBench2.jpg)
>>28830 will do! >>28831 hot. >>28834 I put in as much effort on your images (the first two here) as you did reading this thread.
>>28915 If you ever do make that tumblr, i'd love to see tips on generation and Photoshop morphing.
(266.28 KB 990x1170 BBWChanUpdate3.jpg)
(1.48 MB 3042x3042 bbwChanUpdate4.jpg)
>>26994 >>21689 A couple reworks for the hell of it. >>28932 Well, until that happens, there are a few resources already available. There's a brief morphing guide here >>9226 and I've got an SD guide that I've uploaded here: https://pastebin.com/i8TKvCna It's a little outdated as things have changed a bit in the last year since I wrote it (like the exceptional PS plugin that interfaces directly with A1111 and ComfyUI, now I never leave PS!), but the process/theory will still be useful.
>>28944 shit, formatted the link wrong. The morph guide is on >>/ee/9226
>>28946 oh good lord. >>>/ee/9226
>>28944 Damn... I love when you drop your work in here. Hail to the king.
>>28915 >>28944 this blob girl omfg imagine pumping her throat and watching her feed and fatten on more and more thick, fattening cum
>>28821 Don't mention it - I think the content creation space can be very 'take take take' without giving credit when it's due. With that in mind, I just want to state first off that I expect nothing, especially for free. However, if you do enjoy a puzzle and relish the challenge, I have two examples I would love to see 'come to life'? If you don't, that's fair enough. One is Kitchen Bed by StudioFA - I believe you or somebody else on here attempted to do a later scene in the series. The other is Bunnies First Day by Brasbydesign - I think the scale of her would be mindblowing, but I wonder if the side characters would make it hard to work with.
(418.87 KB 1200x1489 HeavyStudio7.jpg)
(268.67 KB 1000x1500 WideStreet1.jpg)
(519.15 KB 1200x1800 WideHallway3.jpg)
(606.92 KB 1500x1500 BBWChan38.jpg)
(721.24 KB 1785x1905 HeavyRoom7.jpg)
>>29017 Still working on these, it's incredibly difficult to get something that looks good. But progress is being made. >>29012 Hey, I've read those stories too! Ah, to dream... >>28974 Groovy.
(347.61 KB 750x915 IMG_5632.jpeg)
(559.07 KB 750x990 IMG_5633.jpeg)
Can anybody take a swing and try to bring these 2 to life? Artwork by Tonnebelly on DA.
>>29203 the first one is already ai dumbass
>>29208 And yet is still cartoonish. What don’t you stick to something more your speed, like Etch-A-Sketch you fucking moron.
>>29191 Hey! The fourth one over is base on one of my images... very nice, and cool to see!
(888.22 KB 1436x2156 HeavyPool3.jpg)
>>29371 Indeed it is! I meant to link back to it, but now I can't find it.

Delete
Report