CivArchive
    Wizard's Vintage Etchings - Etchings v1
    Preview 14047343Preview 13250987Preview 13250950Preview 13250984Preview 13251018Preview 13250998Preview 13251014Preview 13250990Preview 13251462Preview 13251479Preview 13251571Preview 13251588Preview 13251781Preview 13251802Preview 13251815Preview 13251850Preview 13251906Preview 13251986Preview 13252034Preview 13252039

    Etching art style LoRA that produces mostly monochrome images.

    Use etching, in your prompt to trigger the style.

    Contrast Guide
    dark background for more shadows
    bright background for more contrast
    plain background for more isolated subject

    Play around with your CFG settings, monochrome images usually handles a higher CFG than usual. A higher CFG gives more contrast and a lower CFG will make it softer.

    LoRA Strength Guide
    Strength 0.5 = Less style, more tidy and clean, better coherence
    Strength 0.80 = More style, mid coherent images
    Strength 1.0 = Very stylized, more abstract, more impressionistic



    Other settings:

    Checkpoint: PixelWave 09
    Sampler: DPM++ 2m / DPM++ SDE
    Scheduler: Align Your Steps (AYS)
    CFG: 2.5

    Notice:
    Trained on around 40 high-res prints at 1536px max.res.
    Most of the training images are from early 20th century scandinavia (Anders Zorn), so you might run in to some specific bias from that time-era/location.

    Description

    Etching art style,
    Alternative use triggers: Pencil sketch / Painting / Pastel

    Experiment by replacing the trigger with panting or pencil sketch or omitting it completely.

    Heavily influenced by the etchings of Anders Zorn. So as a bonus it handles thicc nude women bathing and old timey headscarfs

    LORA
    SDXL 1.0

    Details

    Downloads
    1,770
    Platform
    CivitAI
    Platform Status
    Available
    Created
    5/20/2024
    Updated
    9/27/2025
    Deleted
    -
    Trigger Words:
    Etching
    dark background
    bright and plain background
    pencil sketch
    oil painting

    Files

    519217_training_data.zip

    Mirrors

    CivitAI (1 mirrors)