Huggingface for lost Loras: https://huggingface.co/quadratrund
Flux2.0 Klein 9B
Important: improve shadows and lighting,
Flux 2.0 klein 9B has the problem, that it will just overlay the image with cum. if you want to have it in there without looking just copied over, user "improve shadows and lighting" so it gets more baked in. even cum has shadow and so on.
Combine it with my sloppy enhancement lora for ultimate mess.
Strength 0.8-1.3
trigger: m0uth,
general prompts: girl in the image has her mouth open and it is full of cum. she has cum in mouth.
you can even go crazy with excessive cum drooling and so on. then it looks like cum vomiting.
I will be gone, so merry Christmas everyone :)
I boycott early access cause I hate it, but please feel free to tip if it is possible for you.
Creating these loras takes buzz and I created the training images for this lora all by myself. Several hours of creating training pics, photoshop them and refining them.
Illustrious 2.0
150 Trainingpics instead of 76.
Goal is still mouth full of cum without the tongue dangling out.
Trigger: f1lledm0uth,
to enforce: open mouth, cum in mouth, cum filled mouth,
Strength: 0.4-0.8
SDXL 1.0
150 trainingpics instead of 78.
Goal is still mouth full of cum without the tongue dangling out.
Trigger: f1lledm0uth,
to enforce: cum in mouth, cum filled mouth,
To make it more messy: cum drooling, open mouth, bukkake, excessive cum,
Strength: 0.4-0.8
Version 1.0
Trained with 76 pics.
Goal was to create "cum in mouth" without the tongue dangling out. Like a pool of cum in mouth, without a messy face that is blasted with cum.
Triggerwords: m0uthf1lled, mouth open, mouth filled with cum, cum in mouth,
Of course you can go extreme with: excessive cum in mouth, mouth wide open ...etc.
Works best with gehenna checkpoint.
It doesn't work that amazing with minimal anime, cause it seems that understanding what very thin lips are is very hard (see the one sample image). So western, cartoon oder anime with actual existing lips are fine.
Description
FAQ
Comments (4)
power 0,2 + destroy oryginal face :(
Works fantastic, and keeps the character consistency from the image used. Thanks for this, hoping to see more from you.
Any advice for aspiring f2k9b lora creators?
well when i test my loras I always have the same issues: the added stuff looks out of place or the image is super weird but the lora is trained well. So what I learned:
1. if you wanna add stuff (like cum and shit), also prompt: "improve lighting and shadows" - then the cum won't look just pasted over the face but literally as part of the image. cause then the cum also gets "shadows" and is imbedded better.
2. If you don't edit only the person but you also ADD something to the reference image you want to give the AI space in the image. For example my latest Lora with the BBC Deepthroat: If you just have a portrait of the girl that is filling the entire image, give the AI and landscape/horizontal frame to work with. so there is left and right space the add the guy with his schlong. If you keep it in portrait, with will get distorted. You will see, the images with the lora in these showcases are always horizontal but the reference image was always a portrait shot.
3. Currently I am working with portrait stuff. I tried to train an anal-lora with 2000+ images but Flux 2 klein just can't get the girl from a portrait shot into a good doogystyle position or whatever without weird shit happen... currently working on this.
Oh and when you wanna make a lora (your actual question), try to use the automated captioning from civit. It works great, always load up 60 images, captioning them, then the next 60 and so on. Never use the same description on every image. civit will give every image a slightly different caption and with this the lora gets more diverse.
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.