Rosy Haze 柔粉雾光画风 | MOSS-Image LoRA

粉紫色调的柔光二次元插画风：整体蒙着一层淡淡的粉色光雾，线条细、发丝轻，脸颊腮红明显，适合校园、冬日、夜景这类安静可爱的单人题材。

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

• @tokkyu（放在提示词最前面）

训练标注是英文 tag（WD14），推荐用英文 tag 写提示词。MOSS 的画风 LoRA 不写触发词也会生效，写上更稳。

推荐设置

• 不加 DMD：40-50步，CFG 3，euler / simple，shift 5

• 加 DMD：叠加 8 步 DMD 加速 LoRA，12 步，CFG 1，euler / simple

• LoRA 强度：1.0

• 分辨率：约 1MP，例如 1024×1536、832×1216

底模说明：在 MOSS-Image SFT（int8）和 Base（int8）上都测过。SFT 自带画风比较重，会把效果往它那边拉；想要画风更明显，用 Base，或在 SFT 上加 DMD（颜色更浅更亮，最接近原画风）。

提示词示例

@tokkyu, 1girl, solo, silver hair, long hair, wolf ears, wolf tail, winter coat, scarf, standing in the snow, night, street lights, looking at viewer, smile

@tokkyu, 1girl, solo, blonde hair, short hair, school uniform, blazer, sitting on a desk, classroom, sunset, window, holding a book, looking at viewer

@tokkyu, 1girl, solo, black hair, twin braids, yukata, holding a paper fan, summer festival, fireworks, night sky, upper body, looking back

负面：worst quality, bad anatomy, watermark, jpeg artifact, blurry

小提示：

• 团扇、招牌上出现乱码字，是提示词没指定文字造成的，不是 LoRA 的问题；想要文字就把内容写清楚。

• 训练图都是单人图，写单人题材效果最好。

训练信息

• 底模：MOSS-Image Base

• 数据：36 张，2023 年以后的作品，剔除了线稿和带大字的海报

• 训练器：MonadForge-Kohaku（本地 RTX 4070 Ti SUPER 16G）

• rank 32 / alpha 128，学习率 1e-4（前 5% 预热），每步 1 张；e112 = 4032 步，e140 = 5040 步

• 标注：WD14 英文 tag + 触发词

English

Rosy Haze anime style for MOSS-Image. Soft pink-purple tones with a light hazy glow, fine linework, airy hair strands and noticeable blush. Works well for quiet, cute single-character scenes: school, winter, night streets.

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

Trigger: @tokkyu at the start of the prompt. Trained on English WD14 tags, so tag-style prompts are recommended.

Versions: e112 = recommended, brighter and most balanced / e140 = stronger style but darker and duller.

Settings: 40 steps, CFG 3, euler / simple, shift 5; or with an 8-step DMD LoRA at 12 steps, CFG 1. LoRA weight 1.0. Around 1MP, e.g. 1024×1536. Tested on MOSS-Image SFT and Base (int8). SFT has a strong built-in style that pulls the result toward it; for a stronger effect use Base, or add DMD on SFT.

Training: 36 images, MOSS-Image Base, MonadForge-Kohaku, rank 32 / alpha 128, lr 1e-4, 1 image per step; e112 = 4032 steps, e140 = 5040 steps.