Cogecha 焦茶 style LoRA
詳細
ファイルをダウンロード (1)
モデル説明
This is a loRA trained on the style of Cogecha's art
Features
Is easily composable with character loras
Can generate both landscape and characters, though landscape will require the keyword "scenery", and will occasionally have people pop up randomly.
Compatibility
Trained on the NAI model, and is therefore compatible with any anime-styled model such as Anything, Counterfeit, AOM, and PastelMix, etc.
Recommended Weight
I recommend you use the lora at weight = 1.0. If the image fries when composed with another lora at weight = 1.0, it is most likely the fault of the other lora.
How do I prevent frying the image
If your other loRA causes your image to fry, you can prevent this by downloading https://github.com/mcmonkeyprojects/sd-dynamic-thresholding and:
enable dynamic thresholding
set your mimic CFG scale to some low number like 5 or 6
open up dynamic thresholding advanced options
select cosine down as the CFG scale scheduler