Hey Herman and all, after the advice received from Herman, I have decided to recreate and start from square one including redesigning a better model and such. Regarding Herman's last video, does this mean training a LORA is not needed now since he uses three pics to generate the consistency of the vidyz background etc ?