CivArchive
    T2I Image Moss SFT - FLowEdit v1.0 (Test)
    NSFW
    Preview 146177923
    Preview 146177922
    Preview 146177925

    MOSS FlowEdit AutoVision — v1.0 (Experimental)

    Edit your images with MOSS-Image using a mask and a simple text instruction — no LoRA or training required.

    The already-loaded Qwen3.5-9B uses its vision capabilities to describe your photo, then prepares the target description from your request. Just write what you want to change in English or Russian.

    Features:

    • Automatic source and target descriptions.

    • Masked editing: white areas change, black areas are preserved.

    • Adjustable output resolution: longest side, megapixels, exact dimensions, or original size.

    • Source and target text previews.

    • One main workflow with a single Russian / English instruction block.

    Quick start:

    1. Copy the included node-pack folder into ComfyUI/custom_nodes and restart ComfyUI.

    2. Open the workflow and select your MOSS model, Qwen3.5-9B text encoder, and FLUX.2 VAE. Set CLIP type to moss_image.

    3. Upload your image and paint the editing mask.

    4. Choose the output resolution, enter your request, and press Run.

    Experimental release: results depend on the image, mask, instruction, and model settings. Increasing resolution does not guarantee better quality.

    All models download from: Comfy-Org / MOSS-Image

    My recommended text encoder: Qwen3.5-9B Uncensored Heretic — ComfyUI

    My NSFW channel: AI Bob Public
    My SFW channel: AI DEV LAB

    If you enjoy and benefit from my work, please support its continued development using the “Tip” button on my Civitai profile or by making a voluntary donation in USDT.

    • BNB Smart Chain: 0xdCf9846A1B0F2fe32E19dC9CFdDC5194CC13D108

    • TON: UQCeukjBulZw9gni1jv8NLRgtmgHDQ6mUiS2qf3716XobIxA

    • TRON / TRC20: THLU3yu8VpoPqP4bMLCp86FRhoCweJ2Qz4

    Please send only USDT using the matching network. Thank you!
    ------------------------------------------------------------------------------------------------------

    T2I Image Moss SFT

    All models download from: https://huggingface.co/Comfy-Org/MOSS-Image/tree/main
    I strongly recommend using this text encoder: https://huggingface.co/DarkRomeo88/Qwen3.5-9B-uncensored-heretic-ComfyUI/tree/main

    My NSFW Channel: AI Bob Public
    My SFW Channel: AI DEV LAB

    If you enjoy and benefit from my work, please support its continued development by using the "Tip" button on my Civitai profile or by making a voluntary donation in USDT.

    • BNB Smart Chain: 0xdCf9846A1B0F2fe32E19dC9CFdDC5194CC13D108

    • TON: UQCeukjBulZw9gni1jv8NLRgtmgHDQ6mUiS2qf3716XobIxA

    • TRON / TRC20: THLU3yu8VpoPqP4bMLCp86FRhoCweJ2Qz4

    Please send only USDT using the matching network. Thank you!

    Description

    FAQ

    Workflows
    MOSS Image

    Details

    Downloads
    20
    Platform
    CivitAI
    Platform Status
    Available
    Created
    10/19/2026
    Updated
    10/19/2026
    Deleted
    -

    Files

    t2iImageMossSFT_floweditV10Test.zip

    Mirrors