SDXL
v2 is an adjuster/slider model that comes in a SDXL and SD 1.5 model variant. It's trained on front, below and from behind images. It was trained before pony and illustrious models existed but seems to work on many mixes if the weight is set high enough (more than 2 on some models)
Flux.2 Klein 9B
v7a is trained on Flux2-Klein-9B. It was trained as an image-to-image model but also seems to work for text-to-image. The main keyword used for training is "fix her crotch" but it is not required for it to work. To simplify training, unlike v6, this model was just trained with the slit pussy type. I may release other variants in the future.
v7b is a Flux2-Klein-9B img2img model similar to v7a, but trained on an innie pussy with a protruding clitoral hood.
Z-Image
v6 is trained on zimage (and also works with turbo). It's a multi-concept lora. Using "pussy slit" or "slit" will create an innie. Other concepts that seem to work at least some of the time in testing include "clitoral hood", "spreading own vagina with fingers, one hand", "spreading own pussy with both hands", "penis", and "vaginal sex"
v3 is trained for z-image turbo. It's an adjuster/slider model with no keyword needed.
Qwen
v4 is trained for qwen-image-edit-2511. It should handle input images in most styles including anime or furry content. it should also do a better job of drawing buttholes than previous models and also supports images with anal penetration (vaginal penetration not trained). The main activation keywords to use are either "adjust her pussy", "adjust her pussy and anus", or "adjust her pussy, anal"
Chroma
v5 is trained for Chroma. It works best when the keyword "cleft of venus" is used.
Stable Diffusion 1.5
v1 is an SD 1.5 adjuster/slider model. It's trained only on low angle images from the front.
v2 is trained on images from more angles (note there are 2 different v2 models. An SD 1.5 version and a SDXL version)
Description
The SDXL release of labiaplasty v2
FAQ
Comments (26)
The most surprising part of this LoRA is not the end result, but the size of the LoRA. Have you seen the size of SDXL LoRAs? It's like a BluRay movie!
got an error saying 2.0 isnt sdxl
2.0 comes in an sdxl version and a 1.5 version. Are you using the right one? (If you've renamed the files, you can still tell by checking the metadata or looking at the filesize, otherwise, the sdxl one should have sdxl in the file name)
@m99 when using DrawThings, the pruned models are rejected during import. Please provide an unpruned version of the SDXL version, as the currently available unpruned version is showing as SD1.x in both CivitAI and DrawThings.
if you use the lora control extension to trigger it after 10 steps, it barely affects the rest of the image while working as advertised, pretty flawless.
Excellent model for SDXL - it is the only XL one I have got to work
This is the first lora that doesnt seem to work for me, like it doesnt appear in the list and i cant choose it. Any reason why> some help to fix this? using automatic1111
Which version are you using? Auto1111 filters out sdxl lora's if you have a sd1.5 model loaded.
@m99 yeah that was the issue, i also didnt know each model was a separate download, i thought the title had both of them so it would be some sort all in one, fixed it now lol.
A MUST HAVE. period
Getting an error during import of the 2.0 SDXL version using Draw Things. It appears something is noncompliant with this version.
Has anyone been successful in getting the 2.0 SDXL version to load in Draw Things?
Just uploaded an unpruned sdxl version. I think this should work with draw things.
I may be naive but where can I download this?
where is the sdxl version?
Can it do outie when used with negative strength?
works very well! How did you manage that only the pussy is changed? Normaly I got the problme that my lora changes the complete image. But yours keeps the overall image nearly 100% like the original.
I use many tricks to minamise the side effects of my loras and some of them depend on custom training code I'm not ready to release, however a big one everyone can do is choose the right base model.
When training a lora, it will learn what weights in the base model need to change to get the desired output. If you train an anime character of a photorealistic checkpoint, it will learn not only the character, but the anime style itself.
If you want the lora to be trained on one specific thing, the base model should idealy know how to make everything else in your training data except that thing. The best way to chosse this base model is to not choose one at all, but make it yourself.
Looking at your profile take the JustCum model as an example. You could make edits of all your trining data to remove the cum and train a model with the cumless edits then use that model as the base for training the cum lora. That way the lora will learn less about camera angles, lighting, backgrounds, and people in the training data than it would if you didn't make a custom base model.
The base model of your custom base model should either be the official SD weights or a commonly used checkpoint. The custom base model isn't something you release, it's just used for that training set then discarded.
@m99 A brilliant idea, looking forward to the possibility of you publishing a more detailed tutorial in the future.
@m99 very interesting ! Sounds like alot of work... Maybe I will try it with some loras I struggle with. Thanks for that info! Edit: Hah just saw your pussy remover model. So that could be the base model for this model here it seems =)
does not work
sdxl version does not work in easy diffusion:
"You're trying to use a SD1 LoRA model with a SDXL Stable Diffusion model. They're not compatible, please use a compatible model!"
Looks like they check for the presence of one specific block in the lora to determine which version of SD is used
https://github.com/easydiffusion/sdkit/blob/ec209343c69e37ccea8611a540070f6b85af8830/sdkit/models/model_loader/lora.py#L74
The sdxl version of this lora was trained on only a subset of blocks and doesn't contain the particular block they use as a check of if a lora is sdxl or not. I'll keep that in mind for future releases.
Just created an "unpruned" sdxl version that should fix this issue. Let me know if it works.
yep SDXL version still not working for me. would love to have this asap.
Details
Files
m99_labiaplasty_pussy_2_sdxl.safetensors
Mirrors
m99_labiaplasty_pussy_2_sdxl.safetensors
m99_labiaplasty_pussy_2_sdxl.safetensors
m99_labiaplasty_pussy_2_sdxl.safetensors
m99_labiaplasty_pussy_2_sdxl.safetensors
m99_labiaplasty_pussy_2_sdxl.safetensors
m99_labiaplasty_pussy_2_sdxl.safetensors
m99_labiaplasty_pussy_2_sdxl.safetensors
m99_labiaplasty_pussy_2_sdxl.safetensors
m99_labiaplasty_pussy_2_sdxl.safetensors
m99_labiaplasty_pussy_2_sdxl.safetensors
PUS-labiaplasty_pussy_2_sdxl.safetensors
sdxl_vag_standy.safetensors
m99_labiaplasty_pussy_2_sdxl.safetensors
m99_labiaplasty_pussy_2_sdxl.safetensors
Available On (2 platforms)
Same model published on other platforms. May have additional downloads or version variants.