CivArchive
    gemi_minimax_v1 - v1.0
    NSFW

    This is a LoRA of an original character named Gemi. I aimed for a natural, approachable face like a woman you might see on the street,

    Gemi is a fictional character created with AI and does not represent any real person.


    GemiというオリジナルキャラクターのLoRAです。街で見かける普通の女性のような、自然で親しみやすい顔立ちを目指して作りました。

    GemiはAIで生成した架空のキャラクターであり、実在の人物ではありません。

    Description

    FAQ

    Comments (17)

    Hello_World_ARAug 11, 2026
    CivitAI

    This is really good lora. Thank you!!! and I hope you can make more :) / Can we know your traning tool?

    tigertakato
    Author
    Aug 11, 2026

    Thank you! Glad you like it 😊 I used AI Toolkit for training.

    fauxtographyproAug 11, 2026

    @tigertakato Nice work. Was your dataset images and video? I want to try just images...

    tigertakato
    Author
    Aug 11, 2026· 1 reaction

    @fauxtographypro Thank you! The dataset was images only — I didn't use any video. Training with just images works well with AI Toolkit, so feel free to give it a try.

    fauxtographyproAug 11, 2026

    @tigertakato I've been trying all day with the help of claude and I can't even get it to train. I don't get it.

    pijseh2244955fdfdAug 11, 2026· 4 reactions
    CivitAI

    Ito

    brnfd24434343dAug 15, 2026

    May you kee-p making loras!

    tigertakato
    Author
    Aug 15, 2026· 1 reaction

    @brnfd24434343d Thank you so much! I definitely will 🙏

    youmakemeabelieverAug 11, 2026
    CivitAI

    Can you share your AI Toolkit config? Your lora is amazing!

    tigertakato
    Author
    Aug 11, 2026· 1 reaction

    Thanks! It's hard to share the full config in a comment, but the important part is simple: turn off Auto Frame Count and Do Audio, and change Num Frames from 39 to 1. That's it — everything else is the same as a normal LoRA training setup.

    LokitsarAug 16, 2026

    @tigertakato thanks for the advice. I'll try that out. I was having better luck training characters and styles with images surprisingly versus videos but still wasn't getting the results I wanted. I'll try your advice. That said, videos were necessary for training camera movement loras for the obvious reasons.

    ProvenFlawlessAug 16, 2026

    Could you provide some video samples you used for the training? Were they all crop video face shots or a mix of random body shots?

    tigertakato
    Author
    Aug 16, 2026· 1 reaction

    @ProvenFlawless Thanks! I actually trained on images rather than video clips. The dataset includes close-up face shots, of course, but also bust-up and waist-up shots. I also added some shots with more angled views (turned to the left and right) so the model doesn't get locked into a straight-on face.

    tigertakato
    Author
    Aug 17, 2026· 2 reactions

    @GlowingGuardianGirl  出来が良く感じなかったので一旦削除しましたが、再び載せます

    ProvenFlawlessAug 17, 2026

    @tigertakato Interesting. Question: Theoretically if I trained a lora based on a celeb (as you know, H3 has a problem with distance shots--faces become slop unless you increase the steps) could that correct H3 identity even further? As H3 already knows celebs and other people, would that therefore stabilize the face and increase face consistency? Say if I make a lora of tom cruise. Could that conflict or enhance the identity?

    tigertakato
    Author
    Aug 17, 2026

    @ProvenFlawless それほど詳しいことは、私ではわからないですが、理論的には可能だと思います。ただ、有名人の顔の学習はNGですね

    tickaestore799Aug 11, 2026· 2 reactions
    CivitAI

    Never seen this kind of women in the streets in my country. Lol

    LORA
    MiniMax H3

    Details

    Downloads
    645
    Platform
    CivitAI
    Platform Status
    Available
    Created
    8/10/2026
    Updated
    8/19/2026
    Deleted
    -

    Files

    gemi_minimax_v1.safetensors

    Mirrors

    HuggingFace (1 mirrors)
    CivitAI (1 mirrors)