CivArchive
    Pussy sandwich - v0.2
    NSFW
    Preview 684815
    Preview 684816
    Preview 684814
    Preview 684817
    Preview 684813

    Share your images here and leave a like if you enjoyed it! Get exclusive NSFW models at DeviantArt or Booth.pm. You can also follow me on X.


    The LoRA does not have one single trigger tag. Instead describe what you want using a combination of the tags below. Tweaking them will give different effects.

    Main tags: ass, pussy, tribadism

    Penis: 1boy, out of frame, penis, grinding, cooperative grinding

    Other helpful tags: pussy sandwich, looking back, from behind, ass focus, lying, girl on top


    The training was tailored for Illustrious anime models (e.g. WAI-NSFW, Hassaku XL, Prefect illustrious XL). I recommend using at least the quality tag "masterpiece" in your prompt. Recommended parameters: Euler a, 20-30 steps, CFG 4-7.

    Description

    FAQ

    Comments (10)

    RoiD1kanMay 4, 2023· 6 reactions
    CivitAI

    Another great Lora, master! Keep'em coming!

    neonek1232363Aug 29, 2023
    CivitAI

    is there a way to change the appereance of one of the characters?

    i know adding rainbow hair makes the hair different colors, but is there a way to specify which girl has which accessories, hairstyles/colors and stuff?

    psoft
    Author
    Aug 29, 2023

    that’s very hard, it’s an issue with Stable Diffusion not knowing how to identify placement of things.

    but you may be able to do something about it using the Latent Couple extension for automatic1111 webui

    White_FlareFeb 29, 2024· 2 reactions

    I just do it in post with inpainting. Set the scene with the original generation, then inpaint each character with their description.

    jakobhovman259Jul 24, 2024· 5 reactions

    Hello, I use brackets around each individual, including their special hair...example:
    ---
    2girls, (Pia, girl, green hair), (Mia, girl, red hair),
    ---

    and so on for other differences.
    Works most of the time but sometimes SD will make a wild move, however careful we prompt...!
    https://civitai.com/images/21190330

    A1322Jul 26, 2024· 1 reaction

    This is a year old, but regional prompting (a A1111 extension) is the most reliable method. It's got a bit of a learning curve, but it lets you break an image up into sections where each area can receive their own prompt (along with a common prompt that applies to all of the regions).

    So, in this case, you'd probably cut the image into halves vertically, and then you could put "rainbow hair" in one section and not the other, and it would only apply there.

    PressydentOct 25, 2024· 2 reactions

    For Forge you can also use Forge Couple. It's similar to RegionalPrompter but it's working with Forge and Pony.

    jakobhovman259Oct 25, 2024

    @Pressydent hello...I have the Comfy Couple, how can I find this Forge Couple...?
    Greetings

    PressydentOct 25, 2024· 1 reaction

    @jakobhovman259 Afaik it's only useable for Forge AI. If you are using Comfy I think you can't use it. There you can download it from the "Extensions" tab => "Available"

    jakobhovman259Oct 25, 2024

    @Pressydent Thank You , not so great at adding extensions yet.
    Found it via Google:
    https://github.com/Haoming02/sd-forge-couple.git