If you make some cool images, please consider sharing them. That is why I make these. :)
I am the Reindeer Czar, so I created my second LORA, of Red-Nosed Raider from Fortnite. This depicts her really well I think.
Version 2 improves the model very much. No more issue with background, overall more expressive and high quality. 0.5 - 0.7 weight for LORA is good. Euler a and DPM++ seem to be best sampling. Sample around 50 seems good too, more can be better too.
You can add tags like sitting, open mouth, smiling, holding weapon, lightsaber, smirk, candycane, shotgun to use for some dynamic situations.
If a model keeps putting her straight on, facing the camera, add straight on to the negative prompt.
If you like these, please consider a tip, to help me train more models!
https://ko-fi.com/reindeerczar
Deprecated V1 Description:
Strength for LORA is say, 0.4 to 0.65 to get the good result, if you want a background.
It was trained on 5850 iterations, so quite much. Add the tags blue gradient background and simple background in the negative tag to get rid of the background better.