CivArchive
    Preview 31246945

    Example prompt: "A woman's face. Whitish translucent cum is splattered across her face and chest."

    Flexible prompting: see example images.

    Note: "whitish, translucent cum is splattered..." keeps character likeness. If you dont care about likeness you can try things like "her face is covered in milk" or "milky cum" and you might get thicker goo.

    Important descriptors: whitish, translucent, splattered
    Other descriptors: gooey, stringy, thick, wet, milky

    Settings:

    LoRA strength: 1.0
    Steps: 20 - 30

    Goals:

    • Should not impact: ethnicity, age, style, image quality, camera angle etc

    • Small file size (quicker renders)

    • Decent goop

    History:

    V1 (first attempt): complete failure, deleted
    V2: unrealistic
    V3: better goop, but poor quality and too much training person likeness
    V4 (after 17 attempts): better goop, still too much effect over age/ethnicity/likeness
    V5 (after 90 attempts): least influence over subject, but goop can be a little crazy
    V6 (hunders of attempts): slight influence over subject, but better goop than V5

    ⚠️ beta versions will be removed ⚠️

    Description

    FAQ

    Comments (12)

    rollpollution193Sep 26, 2024· 15 reactions
    CivitAI

    I have a suggestion. I'm realizing that the word "cum" has been intentionally "poisoned" in the original dataset (along with other NSFW words). Therefore, whenever we use it, it brings forth that poisoned bias and we have to fight with the model.

    You are going the right direction by describing the quality of it, but I've found better results when I don't include the poisoned token at all. So something like "A white translucent liquid is..."

    1572079Sep 26, 2024

    yeah this is very true

    mawedesign
    Author
    Sep 26, 2024

    interesting!

    mawedesign
    Author
    Oct 7, 2024

    better results for prompting or training?

    rollpollution193Oct 7, 2024· 1 reaction

    @mawedesign I was getting better results for prompting. Also was using in combination with other loras that seemed to be using this technique.

    Don't have data to train this particular concept yet.

    Davros666Oct 15, 2024

    I find that Flux.1-schnell accepts "white goop" in prompts and you get the same effect as the 'warm male fluid emission' you are looking for.

    Why not jizz, spunk, splooge, nut, manjam, nut butter, skeet, knuckle children, load, wad, jerkins, salty shake, cream sauce, jism, spackle, meat mucus, salty milkshake, manwhich, pizza cheese 🤷

    KiefstormOct 25, 2024

    @mikeblackhawkstinks pizza cheese!?

    Davros666Oct 26, 2024

    @Kiefstorm My wife hates getting too much yogurt on her flapjack or accidentally on her shama-lamma-ding-dongs during an eclesiastical whoopsie in the smoosh-spitoon with her favorite creosote...

    mikeblackhawkstinksJan 18, 2025· 1 reaction

    @Davros666 I have no idea what that means, but yet I know exactly what your saying

    WhatTheGuyOct 9, 2024· 6 reactions
    CivitAI

    other loras fail to combine with additional loras like ringags and other stuff that happens with the face. Yours combines very well!

    mawedesign
    Author
    Oct 9, 2024· 1 reaction

    thats good to hear, thank you

    LORA
    Flux.1 D

    Details

    Downloads
    233
    Platform
    CivitAI
    Platform Status
    Available
    Created
    9/25/2024
    Updated
    5/13/2026
    Deleted
    -

    Files

    COF4-beta1_epoch_1.safetensors

    Mirrors

    HuggingFace (1 mirrors)
    TensorFiles (1 mirrors)