fiascos版本unholy (ID: 689139)

Me experimenting...these loras are failures...but, BIG booty butt, sometimes the results can be nice. Some I plan on bettering for suuuuure.

they might be like really bad...you have been warned :p

What you do with them, I leave up to you.

Feel free to use them as you see fit.

UNHOLY:

This was also a test, results came out better than expected so I thought I would share, who knows you might find it useful, no trigger word. It might be over trained due to some experimenting I was trying to uncover. I do plan on bettering it in the future as I quickly put the lora together to gather some better data images.

Words that work: Nun, veil, horror, angel, crown, horns, halo, cross, latin cross, robe, wings, stuff like that.

weight: All depends on the model, some take it well at higher weights, others don't so it's best to experiment.

CINE:

this was a test, but I have been using it for some images and I dig the results, they are subtle but it gives me that blurry background and DOF I am searching for that some models lack along with lighting that I need at times. Plan on bettering it in the future but for now, it does what I need. so I thought I'd share ?

Trigger word: CINE

additional tokens that work: blurry background, DOF, depth of field, fire embers, embers, glowing, backlighting, bokeh

weight: I like to keep this one low, (0.4) especially when I lora stack.

(I actually like to mix unholy and CINE together :p )

be sure to leave a like if it worked for you and you enjoy, helps motivate me to keep publishing my trash ?

描述:

训练词语: