
BlindBox Artstyle
更新了V3版本,调整了眼睛,俗话说眼睛是心灵的窗口,但是目前还是实验性质的,从稳定性上还是推荐v1_mix版本,
如果你喜欢我的作品,可以点个Like,或者请我 喝杯咖啡,
爱来自瓷器,
[机翻英语]
Updated v3 version, adjusted the eyes, as the saying goes, the eyes are the windows to the soul. But it is still experimental, from the stability or recommended v1_mix version.
If you like my work, you can click "Like" or buy me a coffee.
Love from China.
blindbox_v1_mix ?
-
Add the prompt full body,chibi,
-
weigh : 1
-
checkpoint : RevAnimated or other you like
mix PVC material, is good.
?I would appreciate it if you could share with me the results of your use.
blindbox v1
-
Add the prompt full body,chibi,
-
Recommeneded weigh : 1
-
checkpoint : RevAnimated
Other checkpoint may produce fantastic results, try it yourself.
Can be used with other LoRA.
blindbox v2
-
Reduces the probability of a big head (maybe)
-
Lowered network_dim value, smaller file size
-
More Chibi, Less blindbox
v1 version has good blindbox artstyle
blindbox v3
-
调整眼部细节
ーーーーーーーーーーーーーーーーーーーーーーーーーーーーーーーー
盲盒风格的LoRA,一开始是这么想的,
v1版本 还是比较盲盒的,
增加了PVC的质感,更新了v1_mix,效果很不错,建议使用?
增加关键词chibi, full body,
减少负面描述会好一点,也可自行尝试,
经过测试,如果要和别的LoRA一起使用,还是把那些负面关键词加上吧?
推荐使用RevAnimated模型,
跟一些角色LoRA也有比较好的相性,
v2版本减少了大头的概率,更偏向二头身,
适合一些动漫画风了,也许,
有问题可以去B站问,如果我能解决的话?
姑且放上一个没人气的视频
https://www.bilibili.com/video/BV1Ah411G7Zs/
描述:
blindbox v1
-
Add the prompt chibi, full body,
-
Recommeneded weigh : 1
-
checkpoint : RevAnimated
Other checkpoint may produce fantastic results, try it yourself.
Can be used with other LoRA.
训练词语: full body, chibi,
名称: blindbox_v1.safetensors
大小 (KB): 147575
类型: Model
Pickle 扫描结果: Success
Pickle 扫描信息: No Pickle imports
病毒扫描结果: Success