CivArchive
    Doggystyle POV - V1 - Doggystyle POV - zImage
    NSFW
    Preview 112744990
    Preview 112744989
    Preview 112744991
    Preview 112744988
    Preview 112744985
    Preview 112743509
    Preview 112743512
    Preview 112743511
    Preview 112744983
    Preview 112743501
    Preview 113666407
    Preview 113666819
    Preview 113667246
    Preview 113667673
    Preview 113667817
    Preview 113668110
    Preview 113669594
    Preview 113671140
    Preview 113675769
    Preview 113677775

    Trigger is : Doggystyle pov

    Simple Prompting : blond woman, doggystyle pov, in a bathroom

    All you need is the trigger, you can enter the rest yourself and let your imagination run wild. It works relatively well, but you'll have to experiment with the strength yourself.

    I use two ksamplers, with a resolution of 832x1216. For the first ksampler, I use 10 steps cfg 1.5-2.0 euler/normal, then with upscale latent 1.70 nearest-exact. For the second, I use 5 steps with cfg1.5 euler/linear_quadratic with 0.30 denoise. so far, I have been able to achieve very realistic and high-quality photos.

    The strength of the Lora works well with 0.6-1. You have to see for yourself what works best for your purposes.

    If you have any questions, feel free to write to me. If I receive a lot of positive feedback, I will continue my work and will soon create and upload a new, even better model.

    The training resolution is 832x1216 but different formats are possible.

    Description

    FAQ

    LORA
    ZImageTurbo

    Details

    Downloads
    852
    Platform
    CivitAI
    Platform Status
    Deleted
    Created
    12/7/2025
    Updated
    4/27/2026
    Deleted
    12/16/2025
    Trigger Words:
    doggystyle pov

    Files

    doggystyle_pov.safetensors

    Mirrors

    CivitAI (1 mirrors)
    TensorFiles (1 mirrors)