CivArchive
    LightFurryRay - v4.0
    NSFW
    Preview 1
    Preview 2
    Preview 3
    Preview 4

    一种基于光线和颜色强化的LORA模型

    当您使用bokeh,depth of field,overexposure, 这三个关键字时,LORA会在您的图像中加入一种滤镜风格,并最终让它看起来略微鲜艳一些。

    ENG:A LORA Model Based on Light and Color Enhancement

    When you utilize the keywords bokeh, depth of field, overexposure, LORA will incorporate a filtering style into your image, ultimately giving it a slightly more vibrant appearance.


    日语:光線と色の強化に基づくLORAモデル

    bokeh、depth of field、overexposure というキーワードを使用すると、LORAは画像にフィルタリングスタイルを組み込み、最終的にそれを少し鮮やかに見せます。

    韩语:빛과 색감 강화를 기반으로 한 LORA 모델

    bokeh, depth of field, overexposure라는 키워드를 사용하면 LORA가 이미지에 필터링 스타일을 통합하여 최종적으로 조금 더 생동감 있게 만듭니다.

    *:The model is still undergoing testing, and as a filtering-style model, its performance is not yet optimal.

    Description

    4.0 将 LORA 的训练次数增加到了 50 次,效果还不错。权重设置为 1 时可以正常工作,低于 0.6 时出现问题,高于 1.6 时画面会出现破碎。

    In version 4.0, the training frequency of LORA has been increased to 50 times, and the results are satisfactory. When the weight is set to 1, it performs normally; however, issues arise below 0.6, and the image becomes fragmented above 1.6.

    FAQ

    LORA
    SD 1.5

    Details

    Downloads
    605
    Platform
    SeaArt
    Platform Status
    Available
    Created
    2/23/2024
    Updated
    8/1/2024
    Deleted
    -
    Trigger Words:
    furry

    Files

    Available On (1 platform)

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