CivArchive
    Qwen image 2.1 edit super-duper stable NSFW workflow - v1.0
    NSFW
    Preview 144306580
    Preview 144306583
    Preview 144306586
    Preview 144306599

    Qwen image 2.1 struggles with generating NSFW content; at the time this workflow was created, there were no truly decent LoRAs available for this purpose. The model has particular trouble rendering the vaginal area, and this workflow was designed specifically to address that. It employs a dual-sampler setup: the first creates a high-quality base that adheres strictly to the prompt while filtering out low-probability outputs, while the second uses the resulting noise to beautifully render the rest and add fine details. When combined with the right LoRAs and Detailer node, this yields fairly stable NSFW generations—though the occasional anatomical mishap can still occur. I also added a touch of grain and DLSS 5 at the output stage for realism and, well, just because I could, lol.

    Description

    FAQ

    Comments (16)

    lilhu852Oct 1, 2026
    CivitAI

    Missing node: LayerColor: Brightness Contrast
    installing it keeps failing

    lilhu852Oct 1, 2026

    It seems the one from the image works, only the json file has the issue

    utkbec1
    Author
    Oct 1, 2026

    This node installed without errors for me via ComfyUI Manager; if that method throws an error, you can try installing it manually from GitHub. Overall, the node isn't strictly necessary—it adds just a touch more realism, but the results are perfectly fine without it.

    gerard22463889Oct 4, 2026

    Delete it and replace it with Brightness Contrast V2. It will work.

    lilhu852Oct 1, 2026· 1 reaction
    CivitAI

    Could you make a less powerful but lower vram version? Or what settings could be tweaked? AMD GPU: Not enough GPU memory. Try reducing image resolution or batch size and run again.

    lilhu852Oct 1, 2026

    Reducing resolution to 1 Megapixels does wonders

    utkbec1
    Author
    Oct 1, 2026

    What graphics card model do you have?

    lilhu852Oct 1, 2026

    @utkbec1 rx 9070 xt

    utkbec1
    Author
    Oct 1, 2026

    @lilhu852 That’s strange; I also have 16 GB of VRAM but don't have those issues. Try closing all third-party programs, or download https://huggingface.co/abenzerps/Qwen-Image-2.1-Uncensored-GGUF/resolve/main/qwen-image-2.1-UC-Q4_K_M.gguf — to use it, you'll need a GGUF loader instead of the standard

    lilhu852Oct 1, 2026· 1 reaction

    @utkbec1 that worked, also --lowvram seems to help

    markfjohnsonOct 1, 2026· 1 reaction
    CivitAI

    Hey thanks for putting this together! Sorry for the silly question but I cannot seem to find the image edit lora you have in the workflow. I tried to search as listed as well as swapping out the %20... for spaces but still cannot find anything.

    utkbec1
    Author
    Oct 1, 2026· 1 reaction
    45QNa_718Oct 8, 2026· 1 reaction
    CivitAI

    Out of all the ones I tried this one had by far the best result for me

    contexys290Oct 9, 2026
    CivitAI

    What are the options to make it faster?
    First image took like 40min.

    utkbec1
    Author
    Oct 12, 2026

    I think the problem might be a lack of VRAM; in that case, the model gets offloaded to system RAM and runs slowly. On my RTX 5080, generation takes 80 seconds on average—try using the --lowvram flag when launching ComfyUI.

    JIXSOct 18, 2026
    CivitAI

    This workflow works the best. I see that Qwen 2.1 Turbo has been released, do you have any plans to create a version using the Turbo model?

    Workflows
    Qwen 2.1

    Details

    Downloads
    8,668
    Platform
    CivitAI
    Platform Status
    Available
    Created
    9/30/2026
    Updated
    11/20/2026
    Deleted
    -

    Files

    _xX_utkbec1_NSFW_ref2I_QWEN_IMAGE_2.1_Xx~.json