This Lora was trained on a selection of 1200 personal images generated across several models over the last few years.

The images were captioned with wd-eva02-large-tagger-v3, but the lora also works well with natural language prompts.

There are no trigger words.