Hi, everyone this is my first post on the Forum.
I also want to make an AI generated portrait pack mod.
Seeing as you successfully accomplished yours bombasticmori (congratulations, Your pictures turned out great!),
I have a few questions regarding using Stable Diffusion.
I managed to get Stable Diffusion with Automatic 1111 Web UI going, installed the files created by Severian Void and can now create portraits.
But I am still unsure what Model I should use and what settings would yield the best results. How I understand it, Severian Void made two ways of creating portraits:
1. A Standalone Model in CKPT format (which is already a year old) named: "starsectorportrait_v2.1_SD1.5" it also has an suggested Hypernetwork named "HN_ssportrait_v2_1.5_13431.pt" available.
2. A nine month old Lora named "Starsector_Portrait_LORA_1.13.0.e8" which is supposed to be used alongside with Stable Diffusion 1.5 or any other the Model you want, which basically means that If one were to use a Cartoony looking Model one would get Cartoony looking Starsecor portraits.
My Question is, what seemed to work best for you? The Standalone Model or the LORA? Did you use the Hyperntework? And any tips & Tticks would be appreciated.
Kind regards
NoSTs
PS:I am working on Implementing Biological AI into Starsector, basically uplifted animals such as Wolfs, Monkeys and blue Lobsters. This would be the way to crate portraits for them. It works decently well, (see my pfp) but I am unsure.
