I never told you guys, but I'm a bit of a researcher myself, you know. Obscure python scripts from some guy on the internet - yummy yummy.As such, I figured out how to train Anima Lora and trained one!There's not much to it. It's a first test lora based on synthetic SD1.5 Husbando dataset I made a long while ago.Works without tag. Works better with one - 'chargen art'.As I mentioned, it's a quickie - no refunds.

Files with SHA256: 3109fe711a083ca39ce003918fe57351c40d599256fd796fe07c69cadee0d39e