CivArchive
    Norma Hollowell (ノルムー・ホローウェル, 諾姆·霍洛維爾) / (Zenless Zone Zero, ゼンレスゾーンゼロ, 绝区零) - Norma_v2.0
    NSFW
    Preview 136516804
    Preview 136516808

    Norma LoRA (Zenless Zone Zero)

    Training Information
    - Dataset: 40 Images
    - Epochs: 10
    - Total Steps: 2000
    - Batch Size: 2
    - Repeats: 10

    Recommended Model
    - Epoch 8 (Recommended)

    🚨 NOTICE: Please use V2 instead of V1 for this prompt/LoRA.

    ⚙️ Recommended Settings

    • Weight: 0.8 ~ 1.0

    • Trigger Words: nmh

    • Clip Skip: 2

    • Sampling method: Euler a

    • Steps: 28 ~ 30

    • Width, Height: 1024,1024

    • CFG Scale: 5 ~ 7

    🔍 Recommended Hires. fix Settings

    • Upscaler: R-ESRGAN 4x+ Anime6

    • Hires steps: 10

    • Denoising strength: 0.4

    • Upscale by: 2

    Recommended Settings

    - ADetailer: ON

    - Upscaling: 1.5x ~ 2x

    💡 Prompt Example (Base Outfit)

    (masterpiece:1.1), 
    mechanical hat, layered clothes, white high collar, white scarf, split scarf, hanging, sleeveless turtleneck, white turtleneck, covered cleavage, open blue coat, mechanical cuffs, blue bodysuit, garter straps, thighhighs, white gloves, white lace shorts, white boots,
    nmh, 1girl, solo, very long hair, partially braided hair, twin braids, small breasts, white pupils,purple eyes,
    <lora:Norma_v5:0.8>

    💡 Prompt Example (Coat Off Outfit)

    (masterpiece:1.1), 
    ahoge, very long hair, bare shoulders, bare arms, sleeveless, bodysuit, blue bodysuit, garter straps, thighhighs, white gloves, white lace shorts, white boots,
    nmh, 1girl, solo, very long hair, partially braided hair, twin braids, small breasts, white pupils, purple eyes, 
    <lora:Norma_v5:0.8>
    

    🚫 Negative Prompt Example

    (worst quality, low quality:1.4), (lowres:1.2), 3d, bad anatomy, (bad hands:1.2), text, error, missing fingers, extra digit, fewer digits, cropped, jpeg artifacts, signature, watermark, username, blurry, asymmetrical scarf

    Description

    Reduced outfit weight for better flexibility and prompt responsiveness.

    • Batch size: Changed to 2

    • LR Scheduler: Switched from constant to cosine

    • Network Rank / Alpha: Adjusted from 64 / 32 to 64 / 1

    • Scale weight norms: Increased from 0 to 1

    • Other: Minor argument adjustments

    • Expanded the dataset and refined the prompts

    FAQ

    LORA
    Illustrious

    Details

    Downloads
    114
    Platform
    CivitAI
    Platform Status
    Available
    Created
    7/12/2026
    Updated
    7/27/2026
    Deleted
    -
    Trigger Words:
    nmh

    Files

    Norma_v2.safetensors

    Mirrors

    TensorFiles (1 mirrors)