Type | |
Stats | 497 375 |
Reviews | (92) |
Published | Jan 3, 2024 |
Base Model | |
Training | Steps: 6 Epochs: 10 |
Usage Tips | Clip Skip: 2 Strength: 0.8 |
Trigger Words | Navia, blond hair, long hair, frilled hair, blue eyes, neck ornament, corset dress, cleavage, hat |
Hash | AutoV2 248153F221 |
"As the head of the Spina di Rosula, Navia is devoted to helping people of Fontaine, being likened as a "big sister" who checks in on everyone. She has a sharp mind and knack for investigating and a fancy hat and a big chest heart"
- Not so official wiki information
There are some good Navia LoRAs already, but I wanted to make my own and nobody could stop me.
> Main trigger word is "Navia"
> Recommended weight = 0.6 - 0.8
> Recommended additions: blond hair, long hair, frilled hair, blue eyes, neck ornament, corset dress, cleavage, hat.
> Optional additions: gloves, thighhigh boots, high heels, umbrella, sunglasses.
> Works best with anime and semi-realistic models. Not intended for use with photo-realistic ones. The end result will be somewhat anime-like because of the source material.
> Some good things
--> Sitting and standing poses work well with different models
--> Supports both SFW and NSFW generations
--> Other outfits (bikini, maid, etc) also work
> Some known issues:
--> Sitting / Crossed legs poses may generate more legs than universally accepted
--> Depending on the checkpoint, hands can be messy. Use negative prompts or textual inversion (e.g. badhandv4)