First and foremost, i would like to thank like four people for ever supporting my project. It's not perfect and i'm slow, trust me, i know. I must admit, i got lazy, didn't want to start from scratch, and because of that i still have issues with the eyes. That being said, i am still building up a dataset with clearer features, but tagging 1000 images is annoying when the auto-labler kinda sucks. As for this model, think of it as a second draft, i'm learning how to make a checkpoint in a community that likes to gatekeep AI of all things, so bare with me.

However, i am getting the hang of things and i will be experimenting with cleaner datasets as they pop out of the metaphorical oven. I'm trying my best to create a crisp Lora and bake it into a open source base model. I might abandon Illustrious and move towards Flux as Flux users seem to be more willing to share resources.

Free download will come, i can promise that, but give me a while to figure out if it's actually worth keeping uploaded as i need to do some testing.

Please use the recommended settings, or don't, it's your life. I will say, that Adetailer is a must. I like using a high-res fix, some don't, but you can't skip on the detailer. Again, i botched the eyes.