first ver
My training caption dataset is datashit rn










Hello, trainer of Bloons TD6 style LoRA on SD1.5 is back!! This time it's a character LoRA... Not much for a style LoRA on newer models, I'm only looking for a good finetuned model with good adherence (which SD3.5 medium doesn't have right now for anime.).
noobai is so much peakk!!!!
This is a multi-character LoRA, or do I say, tier path upgrades? However, the most working upgrades are 0/0/0 and 0/0/3, while the others are partially working or one of these just doesn't work. The primary tag is tier/x/x/x/mermonkey, but it requires additional attribute tags for it to work better. Fill one of the x with numbers from 0 to 5, leave the others x with 0. Paths like 1/1/1 does not work, it wasnt trained on first tier upgrades anyway. I'll essentially need to train more to get the other upgrades work
For example, popseidon (0/5/0) has alright results (needed prompt tags to get it like the upgrade), but final harmonic (0/0/5) recognizes the other attributes (like the one on the forehead idk lol)
NO LE OCTOPUS MONSTER :c
IIRC the tiers I did not try are 2/2/2, 3/0/0, and 0/3/0
Works so well on artist tag/styles. Have fun trying!!!!!
first ver
My training caption dataset is datashit rn
1. Dieses Modell dient nur Lernzwecken. Urheber- und Auslegungsrechte liegen beim Originalautor.
2. Bist du der Originalautor eines Modells, kontaktiere uns bitte zur Authentifizierung über unsere offiziellen Kanäle. Wir schützen die Rechte aller Schöpfer. Hier klicken, um es zu verifizieren.
