CivArchive
    Preview 137996917
    Preview 137996916
    Preview 137996924

    MiamiJay 👟🌴 | MultiModal Image LoRA

    Premium SDXL x Krea 2 Versions here:

    https://www.kofi.com/tensorvizion

    https://www.patreon.com/tensorvizion/shop

    Base Model

    • Qwen Image / Flux.2 Klein 9b

    Model Qwen Image 2512

    LoRA Weight

    • 0.8 (recommended)

    • Works well between 0.6–1.0

    Sampling Steps

    • 25–35

    CFG / Guidance

    • 4–6

    Image Size

    • 512x768

    Model Flux.2

    LoRA Weight

    • 0.5 (recommended)

    • Works well between 0.3–0.6

    Sampling Steps

    • 24

    CFG / Guidance

    • 6

    Image Size

    • 512x768

    Model Z-Image

    LoRA Weight

    • 0.85 (recommended)

    • Works well between 0.7–1.2

    Sampling Steps

    • 28

    CFG / Guidance

    • 7

    Image Size

    • 512x768

    More base model versions will be added in the coming days, if there's one you want that I don't have message me and I'll see what I can do!

    Prompt Tips
    Describe clothing in detail for the best results. Include materials, colours, footwear, accessories, lighting, and camera style.

    Example Prompt

    <jordans> stylish young man wearing oversized cream hoodie, black cargo pants, premium white sneakers, luxury streetwear, layered silver chains, modern fashion photography, city streets at golden hour, ultra detailed clothing, realistic fabrics, editorial quality

    Negative Prompt

    blurry, low quality, deformed clothing, bad anatomy, duplicated limbs, watermark, text, logo, oversaturated, poorly fitted clothes, cropped, artifacts

    Description

    Trained on Qwen Image 2512

    Thank you to ModelScope & DiffSynth Studio for the training script.

    LORA
    Qwen 2

    Details

    Downloads
    13
    Platform
    CivitAI
    Platform Status
    Available
    Created
    7/27/2026
    Updated
    8/3/2026
    Deleted
    -
    Trigger Words:
    jordans

    Files

    Shoes_c1-st3000.safetensors

    Mirrors

    CivitAI (1 mirrors)
    Civision (1 mirrors)
    ModelScope (1 mirrors)