CivArchive
    ToonyStark Korean Webtoon Flux | Pony | XL | Lightening | 1.5 - xl_v1.0
    NSFW
    Preview 5545473
    Preview 5545489
    Preview 5545498
    Preview 5545584
    Preview 5545491
    Preview 5545492
    Preview 5545487

    **2024/8/19 Flux Lora alpha version is released for this model! Go ahead and give it a try

    Good capbilitiy to print text on image now with the power of flux dev model

    Recommanded CFG for flux version: 3.5-4

    Steps: 25-30

    **2024/8/1 Pony version is released for this model! Highly recommanded to update.

    The pony version is trained based on 4000 high qaulity webtoon images.

    Recommanded CFG for pony version: 5.5-6.5

    Steps: 25-35

    Qaulity tags for postive prompt:

    score_9,score_8,score_7,score_8_up,score_7_up,score_6_up,

    Recomanded negative prompt:

    score_6, score_5, score_4, deformed, distorted, disfigured, (mutated hands and fingers:1.3), pink text,cropped image, speech bubble, english text, monochrome background, furry, loli, poorly drawn, long neck,

    **SDXL Version

    This model is trained based on 300 high qaulity webtoon images.

    Sample prompt:

    1girl, solo, long hair, breasts, looking at viewer, smile, blue eyes, large breasts, brown hair, shirt, cleavage, medium breasts, collarbone, white shirt, upper body, parted lips, off shoulder, lips, adjusting hair

    Neg:

    (worst quality, bad quality:1.2), blurry, bad-hands-5, long neck, text, pink text, white text, speech bubble, dialog box, speech bubble, english text

    Recommanded Settings

    Image size >= 1024

    CFG 6

    Recommand to use Add Detail Lora with 0.2 weight

    Use polyhedron all eyes with 0.2 weight to improve eye quality

    add dialog box, speech bubble, english text, pink text, white text, etc for webtoon text effect.

    Description

    Toony Stark Webtoon XL model is trained on contents of different webtoon scenes

    It can handel people, objects, different poses, multi-people

    Nature language + style tags will give you the best result to match your prompt.

    eg: masterpiece, best quality, 1 girl drink water in a coffee shop, red hair, t-shirt, sunglass, etc

    See more of the posted pictures for sample prompts.

    Offical SDXL VAE is good enough for the job, 768x1024 shall give you some good result, highres fix and adetailer is only optional if you feel you need more details.

    FAQ

    Checkpoint
    SDXL 1.0

    Details

    Downloads
    1,343
    Platform
    CivitAI
    Platform Status
    Available
    Created
    1/16/2024
    Updated
    4/22/2026
    Deleted
    -

    Files

    toonystarkKoreanWebtoonFlux_xlV10.safetensors