It works in illustrious but I like better how it works on pony and illustrious models fused with pony, I think I'll leave it as a second version for Alter~, I should also train a version for IL uwu)/




Metanoia is something very far from its origin but its first version is just a variation of its base and it could end there but its creator is me ...and when I use a lora a lot I end up creating an alternative version trying to correct some errors (actually no improvement, just cleaned some data)~
PD: My idea for today was a concept (it's taking me longer than expected), for now I'll work on it calmly, here's an experiment ~
It talks about something that was practically completely replaced but remains the same as its original version. Metanoia is this, it is practically the same dataset but I changed all the images with signatures and deleted most of the watermarks and urls...and in theory, this version should generate the same result but I have not tested it.
PD: The previous version, extra epochs and everything else I moved to another place, sorry for the inconvenience and enjoy the style ~
It is someone or something that changes too much from what it was before.
In short, this was Lico until I kept changing its entire dataset and it became something totally new, of the original Lico I only own less than 5% but I chose the best of that variable and if you're wondering. Yes, I will train Lico for the last time but I don't think I will be as hasty as before. I really had a bad experience with that project :/
It works in illustrious but I like better how it works on pony and illustrious models fused with pony, I think I'll leave it as a second version for Alter~, I should also train a version for IL uwu)/
1. The rights to reposted models belong to original creators.
2. Original creators should contact SeaArt.AI staff through official channels to claim their models. We are committed to protecting every creator's rights. Click to Claim
