CivArchive
    Pure Beauty V1 - Pure Beauty V1
    Preview 1972861
    Preview 1972863
    Preview 1973063
    Preview 1972864
    Preview 1972858
    Preview 1972856

    Bean Labs is dedicated to delving deep into AIGC technology, restoring lifelike and emotional portraits.

    Pure Beauty V1 is our first self-developed model that excels at faithfully reproducing the authentic forms of individuals.

    The motivation behind creating this model was to eliminate the "greasy" and "artificial" qualities often found in portraits generated by existing models of the same category. Our aim was to tackle issues of homogeneity and craft a model that could better capture the genuine essence of human subjects, aligning more closely with the aesthetic preferences of the Chinese population. Currently in short supply within the market, this kind of model can be rapidly and effectively applied in industries such as beauty and skincare, boasting significant potential and commercial viability.

    Throughout the creative process, we meticulously curated thousands of portrait photographs, carefully selecting hundreds among them. Following multiple stages of post-processing and adjustments to lighting and shadow, we managed to preserve the subjects' skin texture and the natural flow of light and shadow to a great extent. Consequently, our portrait's facial structures and textural qualities stand out prominently in intricate details.

    Join us:

    Discord: Bean Labs

    Xiao Hong Book: Bean Labs

    QQ Community: 789041523
    Twitter: Bean Labs

    Description

    Please enable facial restoration, and I recommend avoiding high-definition restoration.

    CFG Scale: 7

    Iteration Steps: 15-120

    VAE: 840000

    Sampler recommendation: DPM++ 2M SDE Karras, DPM++ SDE Karras, DPM++ 2M Karras, DPM++ 2S a Karras, DDIM (slightly darker light, better structural feel)

    When generating character images, please use the ADetailer plugin. It is recommended to use the models: face_yolov8n.pt and face_yolov8s.pt.

    For generating professional photography images, it's advisable to include the following terms in the positive prompts in addition to trigger words like: Bean_Lab, Pure Beauty, 1 girl, best quality,Studio lighting, Canon EOS-1D X Mark II, 70mm, f2.8, ISO 200. (Besides studio lighting, directional lighting can also be chosen as needed, such as front light, side light, top light, etc.)

    There are multiple camera models to choose from, including Cannon EOS R5, Cannon 1Dx3, SONY A7R III, Nikon D6, Fujifilm GFX 100S, Fujifilm 50S II, etc. Additionally, using wide apertures like F1.4, F2.0, F2.8, portrait focal lengths of 24-70mm, and sensitivity: between 100-200 is recommended.)

    The above information is for reference, and these values can be adjusted based on the desired final outcome.

    Negative Prompts: ng_deepnegative_v1_75t, EasyNegative.

    FAQ

    Checkpoint
    SD 1.5

    Details

    Downloads
    653
    Platform
    CivitAI
    Platform Status
    Deleted
    Created
    8/11/2023
    Updated
    8/9/2025
    Deleted
    8/7/2025
    Trigger Words:
    Bean_Lab, Pure Beauty, 1 girl, best quality

    Files

    pureBeautyV1_pureBeautyV1.safetensors

    Mirrors

    Other Platforms (TensorArt, SeaArt, etc.) (1 mirrors)

    Available On (1 platform)

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