better faces and eyes
better poses
general quality improvement
A first try at making a character LoRA for SDXL 1.0. This is Loona from Helluva Boss, the lora was trained with 100 pictures and e621 tags, so please use them in order to prompt the lora correctly.
As of now this is still in alpha! Details are rough and you might need a few gens to get what you want.
To prompt : Use "loonasdxl" and then all the tags you want. Note that tags such as "solo", "female", "canine", "white fur" and such won't change anything as they have been pruned from the dataset in order to allow for minimal tag prompting.
Weighting this LoRA is fairly straightforward : anything between 0.6 to 1 goes, with a preference for 0.8 weight.
better faces and eyes
better poses
general quality improvement
1- النموذج المعاد النشر هو فقط لأغراض التعلم والتبادل والمشاركة، حقوق النشر والتفسير النهائي تعود للمبدع الأصلي.
2- إذا رغب المبدع الأصلي للنموذج في استلام نموذجه، يرجى التواصل مع موظفي SeaArt AI عبر القنوات الرسمية للتوثيق والاعتماد. نحن نلتزم بحماية حقوق كل مبدع. انقر للاستلام
