CivArchive
    Ideal - Yomi
    NSFW
    Preview 10979679
    Preview 10979681
    Preview 10979682
    Preview 10979680
    Preview 10979683
    Preview 10979684
    Preview 10979688
    Preview 10979686
    Preview 10979687
    Preview 10979689
    Preview 10985054
    Preview 10985062
    Preview 10985061
    Preview 10985052
    Preview 10985053

    Ideal

    Hi everyone, in this page you will find my "new" long term project (ofc i will keep training loras in the meanwhile), the ideal model serie.

    Those are models obtained by merging loras and checkpoints to obtain flexible models, that excells in something.

    Each "Version" of paradise will have a different "use" as objective, so there won't be a classic v1, v2 etc.

    For info on each version check the apposite section in the page.

    This serie will use nsfw models in the merging process, so the models will be able to produce nsfw content!

    Description

    This version of the model is intended to be a versatile one, it can produce waifus and husbandos; safe for work content and nsfw one.

    It is merged with RAM (retro aesthetic machine) and few loras of mine. So it can produce retrò styles aswell. To do that use the tags "1980s_(style)", "1990s_(style)" or "2000s_(style)" at the of the prompt (you might need to put those in round brackets, but the result is pretty controllable, so you can produce the outcome you like).

    As you can see from my posts, i use this model quite a lot when testing, cause i think it's the most versatile one, i've obtained till now. But that doesn't mean this model is flawless.

    Tips and Tricks

    This model works at:

    clip skip 2

    it works best with:

    euler a
    DPM++SDE Kerras

    but it works also with other samplers.

    I suggest:

    between 20-60 steps for euler A.
    between 30-50 for dpm++SDE Kerras.

    Cfg around 7 for standard generations (there are few cases in which u might wanna lower it, and other times when you wanna increase it a little, generally i think you don't wanna go past 11); a general rule of thumb could be, the longer the prompt the higher the steps and the cfg. Ofc you can just keep it at 7 most of the times.

    I encourage you to test it, and share your works. If you like the model you can support me by doing that, commenting,donating buzzs or even reacting to images from time to time. So that i can keep uploading regularly, without needing to wait too much between trainings.

    FAQ

    Checkpoint
    SD 1.5

    Details

    Downloads
    269
    Platform
    CivitAI
    Platform Status
    Deleted
    Created
    4/28/2024
    Updated
    4/22/2026
    Deleted
    4/16/2026

    Files

    ideal_yomi.safetensors

    Mirrors

    Huggingface (1 mirrors)
    CivitAI (1 mirrors)

    Available On (1 platform)

    Same model published on other platforms. May have additional downloads or version variants.