Applicable base model: https://civarchive.com/models/795765/illustrious-xl
a style for illustrious-xl-v0.1
About
style_1:
just add lora, a style.
style_0_test:
I train this way, so I need to add trigger words such as: artist:artist_name,
The effect is stronger when used alone, but the stacking effect cannot guarantee harmony.
(artist:tianliang_duohe_fangdongye),
(artist:jima),
(artist:ningen_mame),
(artist:ciloranko),
(artist:sho_(sho_lwlw)),
(artist:chen_bin),
(artist:ishikei),
(artist:ask_(askzy)),
(artist:wlop),
Some artists originally have in illu, but it may not be obvious.
Description
Applicable base model: https://civitai.com/models/795765/illustrious-xl
Because I know there will be better illu later, I simply trained this lora as a simple attempt\test.
artist:tianliang_duohe_fangdongye,
weight:1,>=1,
The style is effective, and the original base model is not obvious.
About 2010 pictures of several styles were collected and trained after balancing.
It is recommended to use it this way, that is, add the artist's name, which is effective when used alone.
This model originally contains some styles, but some styles are not obvious.
I found that the style words and some characters together\long labels may become inconspicuous.
It is recommended to turn on high-definition restoration.
[artist:artist_name], or artist:artist_name,
artist:ningen_mame,
artist:ciloranko,
artist:sho_(sho_lwlw),
artist:tianliang_duohe_fangdongye,
artist:chen_bin,
artist:ishikei,
artist:ask_(askzy),
artist:wlop,
artist:jima,
Maybe it can be used in stacked strings?
[artist:sho_(sho_lwlw)],[artist:ningen_mame],
artist:ciloranko,
If the style is not obvious, you can: (artist:tianliang_duohe_fangdongye:1.3),
[artist:sho_(sho_lwlw)],[artist:ningen_mame],[artist:ciloranko],[tianliang duohe fangdongye],[artist:chen_bin],


