CivArchive
    Ink Wash Fusion - v2.0 Ink Wash Dream
    NSFW
    Preview 44761798
    Preview 44761810
    Preview 44761847
    Preview 44761848
    Preview 44761849
    Preview 44761942
    Preview 44761945
    Preview 44761851
    Preview 44761825
    Preview 44761938
    Preview 44762046
    Preview 44761946
    Preview 44761947
    Preview 45052221
    Preview 45052257
    Preview 45052267

    Versions of this LoRA are also available on multiple platforms. I post models on multiple sites, including exclusive ones on specific platforms. Follow me on X (Twitter) or Instagram for new model releases, bounties, and more.

    Traditional art fuses with digital style

    Ink Wash Fusion is a sophisticated blend of traditional ink wash techniques—such as shui mo hua, sumukhwa, and sumi-e—with contemporary digital artistry, resulting in a style that bridges the timeless and the modern. This fusion captures the organic flow of ink on paper through soft gradients and natural bleeding effects, evoking the fluidity and elegance of traditional ink washes. At the same time, the heightened contrast and hyper-sharp textures add a surreal, striking quality to the visuals, creating a balance between the delicate and the bold. The style’s ability to harmonize serene minimalism when prompted with intricate details allows it to evoke both the meditative calm of open spaces and the dynamic energy of complex storytelling.

    I created Ink Wash Fusion because I had quite a few ink wash images that did not fit into the style used by Mystic Sumi. I decided to take all those leftovers and train a new model. It tends to be less refined than Mystic Sumi, but also more flexible and modern.

    Usage

    To use the most recent version of the LoRA, use the following settings:

    • Trigger word: dinkwfsn, as in "a dinkwfsn style painting"

    • Other tokens that work well: No tokens are needed to trigger the style and transform most images.

    • Usage notes: minimal prompts will yield traditional, minimalist ink-wash art (e.g. a single sentence of 10 or so tokens). More complex prompts will create images with a more modern, digital look.

    • Lora Strength: A strength between 0.8 and 1.2 is recommended for Flux prompts written in natural language, but pull back to 0.6 to 1.0 for Danbooru tag Flux prompts that usually name more anime-style images.

    For differences in previous versions, see the version notes to the right.

    Description

    Version 2.0 has been retrained at multiple resolutions. Like Version 1.0, it was selected from a slightly undertrained epoch. The result is a dream-like rendering that fuses ink-wash painting with Flux's realistic style. The result is more saturated, vivid colors that maintain the surreal qualities. I like the default strength but you can use stronger settings (strength above 1.0) to bring out more of the ink wash in each image.

    FAQ