No Keyword required in my testing, but im new to this. Use at LoRa strength 1.0 for reasonable results. Here is my first attempt at making a Flux LoRa from about 120 images. This was made using the in-house LoRa maker found on this site. Captioning is new to me but I made sure to focus on capturing her "assets". I made the mistake of using water-marked images in the training data so like 90% of images have something on the bottom corner. Hopefully someone who knows what they are doing will try to make their own, enjoy!

Files with SHA256: 444fa965bea601e05b7600fbdcecd4914be3cf8fe2d69cf6981d978bcb959927