LOTM XL LORA (Lord of the Mysteries)
LOTM XL LORA (Lord of the Mysteries)
- TypeLORA
- ModelSDXL 1.0
- Trained Wordsamonlotm,audreyhall,forswall,gehrmansparrow,kleinmoretti,leonardmitchell,merlinhermes,algerwilson,azikdeath,azikeggers,derrickberg,ebrlotm,emlynwhite,evernightgoddess,lumianlee,medicilotm,mrfool,adamlotm,ccattleya,xioderecha,rosellegustav,bernadettegustav
- TagLORAanime character characters lotm novel character
LORA based on Animagine 3.1 model for 18 characters from the official/fan art for LOTM/COI + some variations. It kind of works with models like 5th tail, but it does not work with PonyXL models, at all.
It can generate:
1) Klein Moretti + 3 of his personas: Merlin Hermes, Gehrman Sparrow, Mr Fool
2) Audrey Hall
3) Fors Wall
4) Leonard Mitchell
5) Alger Wilson
6) Amon
7) Azik Eggers (Both past and present)
8) Derrick Berg
9) EBR (Eternal Blazing Sun)
10) Emlyn White
11) Lumian Lee
12) Evernight goddess
13) Medici
14) Adam
15) Cattleya
16) Xio Derecha
17) Roselle Gustav
18) Bernadette Gustav
It also learned some things like a “Happy Birthday” image style of the official images, as well as some styles.
When I captioned the images, it was like this: trigger word, lotm, 1girl/1boy
So the best way to trigger it would be prompts like this.
It is possible that model can generate copyrighted content, depending on the prompt (usually it doesn’t). I’ll fix it later with better caption and generated images in the dataset.
To generate images I used ComfyUI, which seemingly wasn’t detected. So I truncated my workflow and pasted it here: https://pastebin.com/GGEX27ce