CivArchive
    Art Style:sy4 - v1.0
    NSFW
    Preview 147861102
    Preview 147861105

    Clear Sky 清透二次元画风 | MOSS-Image LoRA

    干净通透的二次元插画风:明亮清澈的大眼睛、柔和细腻的发丝、通透的天空和光感,配色清爽不脏。

    本 LoRA 为个人爱好训练,与原画师无关,非官方。请勿用于冒充画师本人、伪造约稿或商业用途。

    触发词

    • @sy4(放在提示词最前面)

    训练时用的是 tag 写法,推荐按 @sy4, general, 1girl, solo, ... 的格式写。MOSS 的画风 LoRA 不写触发词也会生效,写上会更稳。

    推荐设置

    在 MOSS-Image SFT(int8)上测试:

    • 不加 DMD:40-50 步,CFG 3,euler / simple,shift 5(颜色更饱满、光影更丰富)

    • 加 DMD:叠加 8 步 DMD 加速 LoRA,12 步,CFG 1,euler / simple(更干净明亮、出图快)

    • LoRA 强度:1.0(可在 0.8~1.2 之间调)

    • 分辨率:约 1MP,例如 1024×1536、832×1216

    提示词示例

    @sy4, general, 1girl, solo, white hair, long hair, fox ears, fox tail, animal ear fluff, oversized hoodie, sitting by a window, indoors, holding cup, looking at viewer, smile

    @sy4, general, 1girl, solo, silver hair, long hair, wolf ears, winter coat, scarf, snow, night, street lights, breath, looking at viewer, smile

    小提示:

    • 画面里想出现文字时,在提示词里写清楚文字内容;没写就出现的乱码字,是没指定文字造成的,不是 LoRA 的问题。

    • 示例图的提示词和种子都写在图片信息里,可以直接复现。

    训练信息

    • 底模:MOSS-Image Base

    • 数据:79 张,WD14 tag 标注

    • 训练器:MonadForge-Kohaku

    • rank 32 / alpha 128,学习率 1e-4,每步 1 张,8000 步(每张图约 100 遍)


    English

    Clear Sky anime style for MOSS-Image. Clean, airy anime illustration: bright clear eyes, soft fine hair, open skies and light, fresh colors. Great for kemonomimi girls, school uniforms and everyday scenes.

    Fan-made and unofficial; not affiliated with the original artist. Do not use it to impersonate the artist or for commercial purposes.

    Trigger: @sy4 at the start of the prompt. Trained on tags, so write prompts like @sy4, general, 1girl, solo, .... The style also shows without the trigger, but it is more stable with it.

    Settings (MOSS-Image SFT int8): 40 steps, CFG 3, euler / simple, shift 5; or with an 8-step DMD LoRA at 12 steps, CFG 1. LoRA weight 1.0 (0.8–1.2). Around 1MP, e.g. 1024×1536.

    Training: 79 images, MOSS-Image Base, MonadForge-Kohaku, rank 32 / alpha 128, lr 1e-4, 8000 steps at 1 image per step (~100 passes per image).

    Description

    FAQ

    LORA
    MOSS Image

    Details

    Downloads
    127
    Platform
    CivitAI
    Platform Status
    Available
    Created
    11/4/2026
    Updated
    11/7/2026
    Deleted
    -

    Files

    sy4_e100.safetensors

    Mirrors

    HuggingFace (1 mirrors)
    CivitAI (1 mirrors)