That dazed, just-woke-up look is absolutely irresistible😀!
这没睡醒迷迷糊糊的样子真是让人欲罢不能😀!
触发词(Trigger Word)
fuuchan
简介(Description)
A character LoRA of fuuchan for the Anima model.
Anima 用的「fuuchan」角色 LoRA。
It primarily recreates her head and physique characteristics: short orange hair, purple eyes, and a dazed expression.
主要还原她的头部与身材特征:橙色短发、紫色眼睛、迷糊眼睛。
Base Model(基座)
Anima (anima-base-v1.0). Not compatible with SDXL / Illustrious / Pony / Flux / Krea.
Anima(anima-base-v1.0)。不兼容 SDXL / Illustrious / Pony / Flux / Krea。
How to Use(使用方法)
Trigger word: fuuchan.
触发词:fuuchan。
Also write her fixed features to keep them stable: orange hair, purple eyes.
同时写出固定特征以保持稳定:orange hair, purple eyes。
fuuchan, short hair, orange hair, purple eyes,
Negative:
worst quality, low quality, score_1, score_2, score_3, artist name, blurry, jpeg artifacts, chromatic aberration
There are no fixed outfits in the training set; clothing can be freely specified using tags. However, two outfits are highly recommended: a Japanese kindergarten uniform and a yellow hoodie.
训练集没有固定服装,服装可用标签自由指定。但是还是强烈推荐两套服装:日式幼儿园制服和黄色兜帽衫
Japanese-style kindergarten uniforms(日式幼儿园制服):
kindergarten uniform, Japanese kindergarten uniform, youchien uniform, yellow hat, yellow bucket hat, white stripe on hat, chin strap, white short-sleeved shirt, blue pinafore dress, pleated skirt, suspenders, name tag, chest badge, red ribbon, white socks, black mary janes
The hat may be removed depending on the circumstances:
视情况可以去掉帽子:
yellow hat, yellow bucket hat, white stripe on hat, chin strap,
》》》》这里应该有张示例图《《《《
Yellow hoodie(黄色兜帽衫):
yellow hoodie, long hoodie, oversized hoodie, (plain hoodie:1.5)
》》》》这里应该有张示例图《《《《
Recommended Settings(推荐参数)
LoRA 权重:1.0(0.8–1.2)。
Training Details(训练信息)
Base: anima-base-v1.0.
12 images · rank 64 · dim 64 · alpha 64.
lr 1e-5 · 3000 steps · bfloat16 · epochs 25 (saved every 5).
Notes(注意事项)
Only for the Anima architecture.
仅适用于 Anima 架构。
The trigger carries identity; writing "orange hair, purple eyes" improves stability.
触发词负责身份;写 orange hair, purple eyes 更稳定。
Description
Base: anima-base-v1.0.
12 images · rank 64 · dim 64 · alpha 64.
lr 1e-5 · 3000 steps · bfloat16 · epochs 25 (saved every 5).



