CivArchive
    Qwen-Image-Layered 2512 - ReComposer
    Preview 115074210
    Preview 115073838
    Preview 115073796

    GGUF models

    Qwen-Image-Layered takes a regular flat image and smartly breaks it down into separate, transparent layers (like in Photoshop).

    Key capabilities: infinite decomposition, recolor, replace, revise, remove, resize, reposition

    Each layer represents different parts of the picture or example: one layer might be a person, another a background object, and so on, with their own transparency. This makes it super easy to edit just one thing without messing up the rest.

    You can even choose how many layers to split into (like 3, 8, or more) and go deeper by decomposing a layer further.

    More info: HF

    Apache 2.0 license

    The model is mirrored here for convenience.

    Description

    A fast, lightweight, single-file design tool for compositing PNG and text layers without the need for heavy design software. Built with Vanilla JS and HTML5 Canvas.

    • Create, reorder, and manage multiple Image and Text layers.

    • Drag, rotate, and resize functionality

    • Support for Custom Local Fonts (TTF, OTF, WOFF)

    • Independent Stroke/Outline control (Color, Width, Opacity).

    • Separate Opacity for Text Fill and Stroke.

    • Full suite of filters (Hue, Saturation, Brightness, Contrast).

    • Access to standard Canvas blending (Multiply, Screen, Overlay, etc.).

    FAQ

    Comments (10)

    juanpiDec 25, 2025
    CivitAI

    Very good job. Congratulations 👍

    5817430Dec 25, 2025

    you're welcome :)

    rivdemon1221554Dec 26, 2025
    CivitAI

    quick question before I spend time downloading this - should I be typing in the given text box, what I'd like separated into layers, or is it an automatic process based on its general knowledge base? I'd like to layer specific things such as hair, face, clothing pieces, or accessories.

    5817430Dec 26, 2025· 1 reaction

    It happens automatically based on the number of layers you specify, however, if you want the model to pay extra attention to something in the image, you can prompt it

    rivdemon1221554Dec 27, 2025

    @sweetmax797 ok I've tested constantly and have had zero luck for my use case. It just doesn't understand layering the things I'm after.

    korneliyoDec 27, 2025

    @rivdemon1221554 let me guess, you are after decomposing a 2D character into layers for animation :) yeah ... it doesnt work...

    5817430Dec 27, 2025

    @rivdemon1221554 share the image and tell us what you are trying to do, we might be able to figure something out.

    5817430Dec 27, 2025

    @korneliyo it works on 2D images too https://civitai.com/images/115321563

    rivdemon1221554Dec 27, 2025

    @korneliyo No, I'm trying to separate pieces of a 2d character as mentioned in my OP. QIL's system isn't suited for my needs of separating hair, face, clothing etc.. I'm trying to find a good paper doll method, but so far it seems the only way is to painstakingly cut pieces out in GIMP, because even SAM3 or Florence isn't accurate enough.

    dmaxxjaxx266Jan 15, 2026
    CivitAI

    does this work with forge neo?

    Checkpoint
    Other

    Details

    Downloads
    136
    Platform
    CivitAI
    Platform Status
    Available
    Created
    12/25/2025
    Updated
    5/13/2026
    Deleted
    -

    Files

    qwenImageLayered2512_recomposer_trainingData.zip