This is my first LoRA so I just made this one for personal use
Keep it from 0.80- 1.30 strength
Using Euler a and automatic
I use about 65 steps
As well as the trigger word is "vera" with the quotes
Best to use an upscaler at 1.5x
Additionally, my negative prompt of, "Old, square jaw, high cheek bones, ugly, dark hair, artifacts, teeth, poor shading, cartoon, fat, brown hair, wide lips, thin lips, glowing eyes, blurry eyes, asian, thin lips, natural lips, adolescence, long lips, ??????, text, logo"
Also I recommend using LucentXL Pony by klaabu with this one
Essentially this LoRA was made off similar enough Sora image generations to attempt to replicate the character at least to a point
Description
Much higher step count
Cleaned up some tags
Added slightly more images to dataset from 176 to 208 images
Trained using LucentXL Pony by klaabu instead of base PonyXL
FAQ
Details
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.





