All characters depicted in this thread are fictitious. Any resemblance to a real person, living or dead, is purely coincidental. No pictures in this thread are intended to harm any individual.
No underage/pedo shit.
>>>Guides<<<Realistic AI Guide: https://rentry.co/RealisticAI
General SDXL Guide: https://rentry.co/biglust-training-and-loras
General Wan 2: https://rentry.org/wan21kjguide
Wan 2 for weak GPUs: https://github.com/deepbeepmeep/Wan2GP
>>>Online generators<<<Grok: https://x.com/i/grok/
Meta AI: https://www.meta.ai/
Reve: https://preview.reve.art/app
Previous
>>937234914
>a new thread emerges
it begins again....
WOMEN IN WHITE LINGERIE
WOMEN IN WHITE LINGERIE
>>937241247lol fuckin hell
>>937241077You're a nice bot
>>937241136She looks simultaneously 12 and 40
>>937241316This one her own, no dicks, zoomed out to show more of her sexy body, pls
>>937241136Ha! the AI was like, nope, no real human would have eyes that far apart, so it made her look less like she actually looks, which is mutantish
>>937241321You know, most people don't know the difference between apple cider and apple juice, but I do!
Thanks for the great Lindsay gens
>>937241136whos that? bjork?
i quite do like
>>937241321Here's a little trick to help you remember: If it's clear an yella, you got juice there fellah; if it's tangy and brown you're in cider town!
>>937241411and yet the AI allows Anya Taylor Joy gens...
SHIFTYMINE in WHITE LINGERIE, TITTIES AND PUSSY OUT for science
>>937240959 (OP)we need more of her as Snow white
>>937241399Meant to say, this one, on her own*
>>937241519Sadly doesn't look like her
>>937241136cute as fuck all
>>937241519Bjรถrk is so sexy!
Thank you!
Never seen her before around here.
Pretzel Edition
Taking requests. What i have is below. Feel free to request an addition, but it wont happen soon.
Traditionally Famous:
Lindsay Lohan
Milana Vayntrub
Alexandria Daddario
Alyson Hannigan
Amanda Seyfried
Ana de Armas
AOC
Ariana Grande
Bella Thorne
Billie Eilish
Cardi B
Emma Watson
Gal Gadot
Jenna Ortega
Jennifer Lawrence
Kamala Harris
Kristen Stewart
Madelyn Cline
Margaret Qualley
Mila Kunis
Millie Bobby Brown
Miranda Cosgrove
Natalie Portman
Nick Minaj
Nina Dobrev
Peyton List
Rachel Zegler
Sabrina Carpenter
Scarlett Johansson
Selena Gomez
Shailene Woodley
Sydney Sweeny
Victoria Pedretti
Winona Ryder
I also have an E-Girl, Influencer, and Pornstar List.
>>937241316not enough cum if she just took 3 loads
>>937241582AOC
Miranda Cosgrove
Rachel Zeglar
Sabrina Carpenter
>>937241591SHOE MAMI RETURNS
>>937241591now were talkin
>>937241591Shoe as an elf is actually really Cute
>>937241582Aoc, Scarjo, and Hannigan please and thankyou
>>937241582Millie Bobby Brown
>>937241582Anon, I respectfully request not a gen, but information on what frontend you use and model, perchance?
thanks
>>937241660kill yourself faggot
>>937241582amanda pushing her tits together waiting for a load
oRo39
md5: aa9984c1e392e95db536c6e177930152
๐
back to school
>>937240959 (OP)Fuck yes. I wanna see snow white sleeze fucked in a cheap motel
>>937241582Would love more Lindsay, please
oRo41
md5: 16eb8aff45ed1644dc799762c6db06b9
๐
How do you all think about this? Are you actually using SDXL text2img workflows for client-ready cases, or do you stick to MJ and similar tools when you need ultra sharp, realism, sharp, on-brand visuals?
oRo42
md5: 0488b7f2db4b1d7a26d9414237307252
๐
>>937241815kill yourself faggot
I get the selling points of ComfyUI โ flexibility, control, experimentation, etc. But honestly, the output images are barely usable. They almost always look "AI-generated." Sure, I can run them through customized smart generative upscalers, but it's still not enough. And yes, I know about ControlNet, LoRA, inpainting/outpainting on the pixel level, prompt automation, etc, but the overall image quality and realism still just isnโt top notch?
Ryan Gosling and Hunter Schafer
oRo43
md5: 0fca5e8c0a6f85b45288e7f33b934aaf
๐
>>937241855kill yourself faggot
>>937241622>>937241685Forge Web Ui, BigLust 17. with DMD2 LORA and amateur LORA
>>937241582post the other lists
Ive found that while ComfyUi is insanely powerful, thereโs one recurring pain point that keeps slowing me down. Switching between different base models (SD 1.5, SDXL, Flux, etc.) is frustrating. What about the bird? Each model comes with its own recommended samplers & schedulers, required VAE, latent input resolution, CLIP/tokenizer compatibility, Node setup quirks especially with things like ControlNet
>>937241847Everyone should ignore the trolls and just feast your eyes on this little cupcake
oRo44
md5: 83588c2e80076b2cfbe03975bbf17c7a
๐
Nobody gens men like Ryan Gosling I'm starting to think it's a skill issue
>>937241865you say this like "haha funny troll request haha" but most of the anons here would fuck the shit out of Hunter, me included
>>937241893thank you kind anon
Iโm a relatively new ComfyUI user and mostly work with the most popular SDXL models like Juggernaut XL, etc. But no matter how I set up my SDXL pipeline with Base + Refiner, I never get anywhere near the image quality you see from something like MidJourney or other high-end image generators
>>937241925Why are you implying it's a troll request?
>>937241582Winona Ryder, Jlaw and Meghan Lhane would be a dream. Thank you dude.
>>937241591Hunter Schafer as zelda
Lizzybro, can you gen ashwu on a cute pink bed next? White wooden frame, pastel pink blanket.
On her back with legs up. White rope or ribbons (which ever is easier), tied to her ankles. Which tie her to the wooden frame.
Bottom of her hands/arms to the side of her body. Tied with also the rope/ribbon. Pulling off the screen.
I saw this image of an asian girl on another thread. Something like this would be cool, but with Ashwu completely naked, no glasses, and we can see both her boobs. So legs might need to be a bit further apart to each side.
Pic for general reference of the idea, not for impaiting :)
I was told that I could try to use a latent switch, but I want to use different conditioning and KSampler settings for each preset as well, so a latent switch doesn't seem to cut it either. How do you guys deal with this? Do you use a lot of switches, bypass/mute nodes, or just create a couple of different workflows and switch between them manually?
>>937241954ahhh my European queen. very nice!
I'm trying to build a "master" workflow where I can switch between txt2img and img2img presets easily, but I've started to doubt whether this is the right approach instead of just creating multiple workflows. I've found a bunch of "switch" nodes, but none seem to do exactly what I need, which is a complete switch between two different workflows, with only the checkpoints and loras staying the same. But its inconsistent?
>>937241954You should consider genning Ariel Winter
>>937242041Bend her over, lets see that ass, and a shoe on her butt
>>937241925The same guy who would fuck hunter would also fuck Pedro Pascal, or John Goodman. You know, homosexuals.
>>937241971Thanks, she is an angel
I find it annoying that when I make changes or try things in the workflow, it seems to save all changes and then I don't have a file I can load to get the "original" workflow back. Maybe I'm missing something, but even in the options there is no mention of destination folder for saved workflows and thats weird
>>937241969would love some premo rodrigo
This brings a whole new level of clarity and reusability to building workflows. No more duplicating massive chains across workflows or trying to visually manage a spaghetti mess of nodes. You can now organize your work like a real toolkit. As someone whoโs been slowly building more advanced workflows in ComfyUI, this just makes everything click. The simplicity and power it adds canโt be overstated
Willow Smith
Millie bobby brown
Sadie sink
Wednesday Addams
Emma (Hermione)
Mara Wilson (Matilda)
Brooke Shields
Natalie Portman
>>937241893thank you sir. is it possible to get her bald with no butt plug?
If you havenโt seen it yet, subgraphs are basically a way to group parts of your workflow into reusable, modular blocks. You can collapse complex node chains into a single neat package, save them, share them, and even edit them in isolation. Itโs like macros or functions for ComfyUIโfinally!
https://vocaroo.com/17NjSejdF668
...
>>937242206Heather O'Rourke and Jonbenรฉt Ramsey
Forever angels
>>937242120Could you please show her gaped asshole?
Is there a way to use a remote GPU service to run my locally installed workflows? I understand it would take time to upload the models every time I change server or session, but with a good connection I want to believe it should be better than the alternative?
>>937242134This is amazing
>>937241954Can you do Jaina proudmoor?
>>937242227Margot Robbie or Alison brie please
>>937242049Iโve waited decades to see that pic ..
>>937242063her blonde classmate
>>937242336She IS the pretzelaw!
I've been using remote services for comfyUI like flowt(now closed), runComfy, runpod and more, but overall it's a pain to configure, maintain or simply it's too expensive because they add other expenses like storage
In the latest version of ComfyUI with the menu buttons in the left side, there is a folder icon with saved workflows. When I save a workflow it goes on the list, but where is the actual file?
>>937242294Yes you can, use runpod or vast.ai. don't you feel bad I actually gave you good information? Now stop.
>>937242336hunter schafer like that
>This thread is sponsored by:
>Schafer Schnapps!
>Taste the meat, not the heat!
>>937241815Squatting, legs wide apart (to see granny pussy), peeing and shitting.
>>937242416Shailener next?
>>937242415She is the princess of my heart
It has also helped me learn a lot but I am a bit stuck on how to proceed. Specifically, I have been running into issues with how to properly create high quality emotions from a single source portrait. In my workflow I've created two versions for how I could approach thisEven with low denoise it seems to modify the subject a lot. Does that make sense?
>>937242415that bottle should have red or blue liquid, and label should say hyrule
>>937241582Natalie Portman for the love of God
Is there a smarter way or workflow to generate consistent multi-character images using all three of my LoRAs together โ ideally without relying on face swapping afterward? Using flux now, but Iยดm open for other suggestions as well. I'm also uploading a reference image of the three of us, if that helps! Any tips or experiences would be really appreciated
bjor
md5: ca370135529b65504ffc74d5e2b8294f
๐
>>937241547>Sadly doesn't look like herReal, for reference
Iโve trained custom LoRAs for each of them using Flux. Right now, my workflow looks like this: I generate an image using the robot's LoRA and just two random male characters. Then I manually do face swaps to replace the random men with my two custom-trained male models
I have more complex questions, mostly about the SDXL base-refiner modules, but I'll wait on them while I look at other workflows to see how the community is doing things. Again, the ability to import other workflows is a huge strength. Currently I'm working on a project where I want to generate images featuring three consistent characters: two men and one robot. Doesnโt that sound exciting?
>>937242586christ, she's hot
Emma Watson,
damn the 6-month ban
>>937242657I miss Joker foreals foreals
I had some questions about getting an XYZ type interface, but I'm already finding some nice workflows I can modify. Are there any helpful extensions to get an XYZ? It's the only script I really use in A1111. I'm a bit of a prompt purist and try to make that perfect instead of inpainting through it, but ranges and grids really help with that
>>937242593ARE YOU GOING TO FUCKING TAKE MY REQUEST OR NOT, YOU MEGA-AUTIST ROBOT?!
>>937242715With that attitude?
>>937242733breaking global rules
you need to keep that shit in MLP
do it again. banned
Any gallery resource for workflows / jsons that I can browse? A place where users can share theirs? Civit has some but it's a bit messy. I assume ImageScale has been replaced by Upscale? Can't find it anywhere. I was gonna use it to reduce 4x ultrasharps, because they're, like, really big. I have questions about how to upscale, but I will research workflows first. Do you think this is a solid approach to success
>>937242741Christina Hendricks
>>937242741even in AI you can tell she cant act
>>937242758lol what was it? whats MLP?
How do I drop old generation data into a new workflow? How do I work with an old PNG without overwriting the new UI of modules? Like putting RealisticVision DPMPP SDE Karras 512x512 CFG 7 seed 1234567 "photo of a ninja cat fighting crime" that I made in ComfyUI a month ago into a fancy new workflow to make some tweaks?
>>937242831Unironically has more acting chops than Zendaya
>>937242858my little piggies
>>937242586just need more Shailene, in any way we can get her
>>937242834Rip to the greatest celeb ai troll
Looks suspiciously like my gf's younger sister... hmm...
>>937241582Greta Thunberg pleaseeeee
what is dead may never die
>>937242872this is the best trolling in the entire thread of retards
I go now. I'll be on again later
I love how simple the ComfyUI install was. Love the interface. And looove how easy it is to reproduce other people's work to see how they did it and learn from them, something of which is very difficult in A1111. Where is the bird? This will likely be my primary app for Ai art going forward, but I have a few questions that I can't answer anywhere else. So I thought this thread of master bakers might be able to help
>>937242902you made this? who made this
>>937243072Noooo, this is an original artwork by the Master
I would never take credit for his works
>>937243101who is the master?
just some thread vintages
>>937242858my little pony. it was a picture of Ryan kissing a pony
velma
md5: 32444a93dd220558e352732dea94cf52
๐
Does attention/emphasis (prompt:1.5) work? I've been elbows deep in Automatic1111 since Spring and having a blast. But I'm always exploring better ways to do this. Ever since I saw that little ComfyUI avatar in the SDXL live chat last week, I had an interest in checking out the app, especially since it was so stable at launch but I must be doing something wrong?
>>937243165Blasphemy! He would never!
>>937243047Thanks for the great gens
>>937243215If only.....
We never realized how good we truly had it
Been watching Firefly again, would love to see any of these celebs show up
Jewel Staite
Morena Baccarin
Summer Glau
How do I learn about specific tasks for my workflow? Most workflows and tutorials I've found look so complicated and I cant' figure out certain elements (like upscale or previews) without learning a whole other workflow. I asked my bird pals but they didnt know either. How do I "translate" my workflow into a1111-prompts? Like how do I figure out the denoising strength for the refiner?
>>937243251No young Christina?
>>937243268The princess we needed
>>937243268her as zelda looks pretty cool
the world was not ready
they casted someone else. was announced today
How do I add a second or a third LoRA? Do LoRAs need trigger words in the prompt to work? And what about the bird? Is some sort of hires fix necessary if I'm generating 1024*1024 images or other image sizes that SDXL was trained on? I understand it was useful for images bigger than 512*512 with 1.5 so I'm wondering if it's still needed
>>937242582>>937241519This is bjork??? Thought it was that one nerdy kid from the 90s just gender bendered
>>937243246It was fine up to two hours ago, bretty fuggin gewd actually. Though if you're into obese ancient women and Ryan fags, this is a good spot!
Roll with the punches, we'll get through this minor bump in the road!
6436
md5: 804ccffc62ff3b66cd9ef48e4896b2a9
๐
>>937243246Celeb ai will never be this good again....
>>937243047You missed portman
Hello! I'm trying learn how to use ComfyUI and I'm struggling to find a good starting point. I've set upa simple workflowwith model+refiner+lora and it works (I think). Now I want to learn how to add elements and I'm not sure how. Is inpainting and img2img possible with SDXL right now and if so, how do I learn how to set up a separate workflow for both? Is there some sort of "put this noodle here and this one there" kinda thing going on?
>Dare ye enter my magical realm?
>>937243274Haha definitely would take her, but I at least see some of her here
oro45
md5: d9f2576ce6443aaaf560c97fe8b73677
๐
cursed thread
goodnight trolls
Is there a way to feed it data so that the AI combines it...say I want it to generate a character. Would it be possible to feed it the pictures of models/actors and have it generate a new image based on those? I know how the people look in my head but the only way I could describe them is by using pictures of people that look like them as I can't draw for shit
oro46
md5: 23e3b95ec26fbfc8e9afad5ad522819e
๐
>>937243532Always up for an adventure
035
md5: ed15d8a92cc9c0c73abecb7e65d9a3a9
๐
Thanks for the cumny Miranda last thread
>>937243635>>937243561that's nice actually. nude but sophisticated
>>937243520No, you're mkultra
I am pretty new to ai image generation.I want to use stable diffusion online and for free. I'm fine with logging in as long as the service is free. I want to use multiple models likeStable Diffusion, Pony Diffusion, etc with availability to use LORAs. So is there any site available to use multiple models for free?
>>937243687who the fuck is that?
>>937243268>>937243301I think 5 or 6 years ago, people would have been cool with the cast. know some rich guy they would pay Hunter to play zelda in their own version.
Flux Kontext can do great stuff, but when it comes to people most output is just not usable for me. When people get smaller, usually about the size that a full body fits to the 1024x1024 image, especially the head and hair start to show artifacts looking like a too strong JPEG compression
>>937243740>Pajeet makes bots compliment each otherMight be the saddest shit I've seen all week
Ok, some img2img refinement might fix that. But when I do "bigger" edits, something Kontext is really made for, it gets the overall anatomy wrong. Heads are too big, the torso is too small.
>>937243809We can kiss next if you wanna jerk off
>>937243809No Ken
No opinion
I'm getting mixed results with Kontext (FP8 scaled). I'm inputting two images and trying to put Keanu Reeves on the surf board. Kontext only accurately recreates the Keanu's face when the resolution is 1024 x 1024 but then the rest of the image is changed.
>>937243714Does it suck to be a newfag
>>937243704you can try tensor or seaart
>>937242748Once again... are you him?
Using similar or exact resolutions as the surfboard image yields poorer results but the composition of the background is better. Also the water looks terrible and looks cartoonish and has a lot of pixelated-like artifacts. This tends to happen a lot in general with images I create, but it's particular bad with water. Hopefully someone will notice something noobish I'm doing wrong
>>937243672More Francesca please
>>937244020Thank you, very kind. One of the rare good anons
>>937243738i think Bella Ramsy ruined any future castings fans dont like lmao
>>937244061i dont even know anymore
>>937243687no I'm DaddyRyan
>>937243537Gaddam that's a dream come true
>>937243532More Hunter cock!
>>937244127I'll make it simple for you: You're not the bot posting grannies, soooo...
... I wAsN't TaLkInG tO yOu!!!!!111
Iโve been messing with AI models and image gen since late 2022. Back then, everything already felt like magic, but it was kinda hard to actually gen/edit images the way I wanted. Youโd spend a lot of time inpainting, doing weird workarounds, or just Photoshopping it by hand
>>937241136>>937241519>>937241899>>937243537So based. Bjork's internet defense force is on full damage control to shut this down
And nowโฆ we can literally prompt edits. Like, โoh, I want to change this partโ and boom, the model can just do it (most of the time lol). Sure, sometimes you still need to do some manual touch-ups, upscaling, or extra passes, but man, the fact we can even do this locally on our PCs is just surreal to me
>>937244272I gotta protect my queen !
>>937244115true. Hunter has the face for it. Like said i think some guy is going to pay her to play his version of zelda
>>937244272that's bjork? needs a better lora lmao
>>937244323Yeah idk if you knew this but bjork actually has that fetal alcohol syndrome look IRL
>>937244245MAKE MY REQUEST, YOU RETARDED BUCKET OF BOLTS!!! RIGHT HERE!!!
>>937242420
>>937244364Thank you, David Tennant
75
md5: c967d2dad28aa3dda2f40e34e3882ae5
๐
take a good look at the last 20 mins folks, this is what they want this place to be
Fun and interesting at least
Not just muh porns
>>937241136is the Bjork Lora shared?
before i forget, just wanted to say thank you LizzyBro! for the gens :D
always doing the lords work lol
>>937244353>blame the art not the artist!
>>937244429We love Kristen though
>>937244429You mean kstew spam?
>>937244429I see fat clowns, old bitches,and an Alfred E. Neuman Princess Peach...
>>937244429THIS IS THE FUTURE THEY CHOSE
THIS IS WHAT THEY TOOK FROM YOU
THIS IS HOW THEY PUNISHED YOU
>>937244429and this is what this place needs to be
where is the session for prime cels
>>937244429People like you are heroes
More Kris10
Save ussssssmgpr8
>>937244656kris10, portman, watson, oretaga, mbb, sophie, sadie, anya and more lol
>>937241969Yes. It's the good good.
>>937241582Marisha Ray
Laura Bailey
Ashley Johnson