FiosFragileEarth

Welcome to my 15th loRA! A Model trainined using 128 images created in Bing Co-pilot Dall E 3

This LoRA enables imagery of Planet Earth portrayed in diverse ways—celebrating its beauty, inspiring awe, & highlighting the urgency of conservation & protection for all species

The core focus was on images featuring a young woman holding a globe, surrounded by animals, waterfalls, & fairies or other spirit-like guardians. Ironically, these proved to be the trickiest to render cleanly in Testing, often resulting in messy animal forms or anatomical quirks.

The dataset also included diverse scenes: animals with Earth backdrops, seasonal globes, waterfall gowns, Earth deities, whales, dinosaurs, owl eyes, Jellyfish, nature-tech conflicts & a lot more! All of which appear to render beautifully!!

The LoRA seems to have wide-ranging use & excels at generating detailed, beautiful women.

The lora does struggle with bad anatomy so patience may be needed.

Seems to work well with other loras.