CivArchive
    Cum XL - v1.1
    NSFW
    Preview undefined
    Preview undefined
    Preview undefined
    Preview undefined
    Preview undefined
    Preview undefined
    Preview undefined
    Preview undefined
    Preview undefined
    Preview undefined
    Preview undefined
    Preview undefined
    Preview undefined
    Preview undefined

    ⚠Please read the release notes for each version!

    ⚠Gallery images for v1.0 were generated with various base models (Realvis, Juggernaut, Zavy Chroma, others).

    ⚠Gallery images for v1.1 were generated with only the base SDXL 1.0 model with 0.9 VAE.

    The bizarre gallery photos are intended to showcase model flexibility.

    I created this LoRA mostly as an experiment to try out a new way of training introduced to me by micvrotom. It uses color masking and prompting to train. Please check out his write-up located here if you are interested. Please check the files if you are interested in the dataset used to train. On version 1.0, I've noticed is that generating the color associated images look a little better than just images by themselves. It might work better as an inpainting model. Version 1.1 improves on this.

    This model can generate both the side-by-side image with the color mask, as well as the individual images, for example:

    • Color-associated region in two identical side-by-side photographs of a naked woman laying on a bed with cum between her breasts and on her stomach. Trimmed pubic hair. The magenta region is cum.

    V1.0 Notes

    Initial test of using color guided training. Only trained on side-by-side images. Released the last epoch of training.

    V1.1 Notes

    Started a new training but added in single images with their own captions to the dataset. Released the last epoch of training.

    Description

    This is trained on the same dataset from version 1.1, but single photos were added instead of only the side-by-side images. This version may be over-trained, so try dialing back the weight depending on what base model you are using. Here are some sample captions that were used during training:

    • Photo of a naked woman with blonde hair. She has small breasts. There is cum dripping off of her mouth and chin and on to her chest. She has hard nipples.

    • Photograph of a woman with dark hair and an open mouth. She is pushing her breasts together with her hands. Her nipples are pierced. She has cum on her right breast. She is in a bedroom and sitting on a bed. She has medium sized breasts.

    • Photograph of a naked woman laying on a white bed. She has cum on her small breasts and inbetween her breasts. Her nipples are hard. She has light brown hair. Her mouth is open.

    • Close up photograph of a woman's pussy. Trimmed pubic hair. Cum is dripping out of the vaginal opening and onto the bed.

    • Close up photograph of a woman's face covered in cum. Cum on her cheeks. Cum on her lips. Cum on her chin. Shoulder length dark hair. Wearing eye shadow and dark eye liner.

    You can download the dataset and see the captions and images for yourself.

    FAQ

    Comments (1)

    1212449Feb 18, 2024· 2 reactions
    CivitAI

    when using "messy-pov" + "cum" there is no visibly difference in images when using your lora or not. uploaded 2 batches for you to see if someone can find a difference

    LORA
    SDXL 1.0

    Details

    Downloads
    5,781
    Platform
    CivitAI
    Platform Status
    Available
    Created
    2/17/2024
    Updated
    5/13/2026
    Deleted
    -
    Trigger Words:
    woman
    cum
    chin
    cheeks
    mouth
    chest
    breasts
    stomach

    Files

    cum-xl-1.1-dataset.zip

    Mirrors

    CivitAI (1 mirrors)

    cum-xl-1.1.safetensors

    Mirrors

    HuggingFace (1 mirrors)
    CivitAI (1 mirrors)