This is a lora made from NIji sweet spot v4 + 748cm style output.I'm still learning how to create a proper dataset for trainning, so this will be experimental version.
BEST use with other than ANIma base01
Experimental version
V1.safetensors