Version 2 is overall big improvment. Simply prompting for Sombra will suffice for whole look, except for optional lipstick.
Suitable strength range is 0.6-0.8. Triggers for paticular things, in case you still need them, are:
Outfit - bodysuit, coat, trench coat, turtleneck, high collar, legwear, gloves, short sleeves
Accesories - jewelry, stud earrings, hologram, holographic interface
Makeup - makeup, mascara, eyeshadow, eyeliner, long fingernails, nail polish
Gun (really wonky, not trained for it) - holding gun, weapon, submachine gun, finger on trigger
Making and testing LORA's takes quite a bit of my time, so I would appreciate if you bought me a coffee.
Trained on Anything V4.5.
Samples are 512x768, not cherry picked, without HighRes.fix, using OrangeCocoa 50/50 Mix , with OrangeMix VAE (for version 2), and using EasyNegative embedding. Inpaint examples are txt2img with 1 single inpaint done on eyes.
Obsolete. Version B - dataset of 50 images with deepbooru tags. sombra \(overwatch\) can be used instead of trigger words.
Go ahead and upload yours!
Your query returned no results – please try removing some filters or trying a different term.