Making models can be expensive. Do you like what I do? Consider supporting me on Patreon 🅿️ or feel free to buy me a coffee ☕
This was requested by a user, but it was already in my backlog so I just bumped the priority. Came out pretty good. Didn't expect to get also Pneuma, as it was just in maybe 10% of the dataset (and she is definitely less accurate than the other 2, who came out perfectly).
Triggers are pyra \(xenoblade\), mythra \(xenoblade\), and pneuma \(xenoblade\). As for the weight, I don't think it's wise to go over 0.5 (check the matrix at the end of the gallery), sweet spots seems to be between 0.4 and 0.6.
I could also make this work with latent couple and controlnet to get both girls in the same image: https://imgur.com/a/XMIYH2I (pnginfo: https://pastebin.com/qmd8xP0q).
I'm also uploading the 42 epoch version because it rarely gave better results, so why not. I've added comparisons in the v42 tab.
How to use LoRA's in auto1111:
Update webui (use git pull like here or redownload it)
Copy the file to stable-diffusion-webui/models/lora
Select your LoRA like in this video
Make sure to change the weight (by default it's :1 which is usually too high)
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
