CivArchive
    3DPonyVision - v1.0
    NSFW
    Preview 13767687
    Preview 13767563
    Preview 13767576
    Preview 13767579
    Preview 13767651
    Preview 13767658
    Preview 13767696
    Preview 13767714
    Preview 13767739
    Preview 13767755
    Preview 13767773
    Preview 13944670
    Preview 13944729

    A simple merge of several PonyXL models in an attempt to match the style of my base SDXL merge, a semi-realistic 3D Disney cartoon style with sharp details and vivid colors. It isn't quite as good as I'd like it to be, and base resolution faces are pretty terrible, but I'm getting good results with hiresfix + adetailer/face detailer, which are highly recommended. A simple ComfyUI workflow including both was used for all the examples.

    I uploaded the workflow separately, you can find it on my profile and in the suggested resources.

    Recommended settings:

    Resolution: 812x1216, 1024x1024, 768x1024 (for slight speed boost)

    Steps: 30-50

    Sampler: Euler a (any should be fine)

    Scheduler: Karras (AlignYourSteps scheduler used in examples)

    CFG: 5-7

    Hiresfix: 1.5x-2x using any upscale model (4x_foolhardy_Remacri used in examples)

    10-20 steps at .3-.45 denoise

    (examples use 1.5x, 12 steps, .4 denoise)

    Adetailer/face detailer is HIGHLY recommended.

    Please experiment further as I didn't change these settings much.

    Recommended starting positive prompt:

    "(score_9, score_8_up, score_7_up:1.1), score_6_up, score_5_up, score_4_up, (semi-realistic disney cartoon, volumetric lighting:1.1)"

    Recommended starting negative prompt:

    "score_4_up, score_5_up" (I used this for the examples, but I'm not sure it helps at all. Please experiment more, I haven't tested prompting very thoroughly.)

    The models used for the merge are listed in the suggested resources section, please check them out as well.

    Description

    FAQ

    Comments (6)

    PolygonMay 28, 2024
    CivitAI

    How's the responsiveness to the prompt, though?

    Zen1
    Author
    May 28, 2024

    @Polygon From my testing it seems to follow pony/booru tags relatively well. The prompts are included in the example images' details and I'd say they match up for the most part, give or take a few concepts as usual, especially for the more complex prompts.

    AldaDiazAug 21, 2024· 1 reaction
    CivitAI

    hey, very nice work! ..i noticed you have "score_5_up, score_4_up" in positive AND in negative as well.. i guess it is intentional since you wrote it as recommendation too, but i mean doesn't it outrule one another?

    Kevin0777Sep 10, 2024· 1 reaction
    CivitAI

    Can I use this for my NSFW games / Comics?

    VelouraNov 18, 2024· 1 reaction
    CivitAI

    damn this is probably the greatest pony model beside Cyberrealistic for prompting. You can get close to perfect gens with little work and effort.

    goblinweeaboo901Nov 10, 2025
    CivitAI

    The results are near perfect compare to other models that gives inconsistent limb orientation and smudged faces, Probably my favorite model to use, Good job seriously

    Checkpoint
    Pony
    by Zen1

    Details

    Downloads
    2,993
    Platform
    CivitAI
    Platform Status
    Available
    Created
    5/27/2024
    Updated
    5/13/2026
    Deleted
    -

    Files

    3dponyvision_v10.safetensors

    Available On (2 platforms)

    Same model published on other platforms. May have additional downloads or version variants.