I absolutely wasn't happy of my last try, so... Another one, yeah.
Fourth and last version of the LoRA: quite beautiful output, even if it's a bit... wild, in saturation.
If you have problems with it, downgrade to 'chenkin sin pr'
Description
Latest version trained with NAI chenkin
Probably the best one. A bit weak, but quite stable: in the test I cranked the weight up to 2.5 without bad consequences, so go for it