CivArchive
    Kyhu artist style - k-y-h-u (old pen name of iafhy) [MiniMax H3/ANI/IL/KR2/ZIT] - zit_v1.0
    NSFW
    Preview 116672079
    Preview 116685290
    Preview 116675474
    Preview 116673151
    Preview 116674938
    Preview 116675321
    Preview 116675498
    Preview 116675539
    Preview 116675615
    Preview 116675688
    Preview 116680924
    Preview 116688345
    Preview 116690017
    Preview 116690697
    Preview 116691337
    Preview 116692945
    Preview 116693049
    Preview 116693109
    Preview 116693739
    Preview 116693997

    This style lora is trained on images by KYHU, the old pen name of artist IAHFY. The dataset mainly consists of images of Korra and Asami, but it also works good for other characters. The images have monochrome or low detailed backgrounds. So this lora is best suited for generating characters without detailed backgrounds for Illustrious. Since Anima seems to be better at generating detailed background, you can try if it does better there, but I haven't really tested it. Works nice with additional style tags like "monochrome", "sepia" or "sketch". All versions are trained with the same dataset and captions (except for the Krea 2 and H3 versions). H3 LoRA seems to work with simple prompts, but may fail for more complex ones. If you want to support the artist, you can find options here.

    trigger (minimax h3): kyhu style image sketch animation, thick lines,
    trigger word (image generation): kyhustyle
    additional style tags: monochrome, sepia, lineart, sketch
    character tags: asami sato, korra

    If you like the model please leave some reactions on the showcase images or videos. Also I would like to see the images or videos you generated. So if you really like the model please click "Add Review", leave a like and add some images, so I get notified you did. If you like you can also write a short review.

    Description

    This is the first ZIT LoRa I am publishing. It is trained with the same dataset and same danbooru tag style captioning as the illustrious version.
    It can be used with tag style prompting, but I noticed that natural language prompts or mixed prompts seem to work better with Zimage Turbo. So using "detailed drawing in kyhustyle of" or "digital art in kyhustyle of" at the beginning of your prompt followed by your main subject should be used instead of just the trigger word "kyhustyle".
    I recommend a weight of 0.5-1.0, but at values close to 1.0 the bias towards korra gets pretty strong, since she is in most of the images in the dataset.
    For the showcase pics I used an fp8 quantized version of base Zimage Turbo, so they look a bit different from the illustrious showcase pics using a finetuned checkpoint.

    trigger words: kyhustyle, korra

    FAQ

    LORA
    ZImageTurbo

    Details

    Downloads
    119
    Platform
    CivitAI
    Platform Status
    Available
    Created
    1/7/2026
    Updated
    10/27/2026
    Deleted
    -
    Trigger Words:
    kyhustyle

    Files

    kyhustyle_zit_v2.safetensors