Sign In

F NoobAi (Star Wars: Visions - The Village Bride) | LoRA

843

Not recently loaded

Download

1 variant available

SafeTensor

54.77 MB

Verified:

Type
LoRA
Stats

121

843

303

Reviews
Published

Jan 18, 2025

Base Model
Training
Steps: 1,472Epochs: 16
Usage Tips
Clip Skip: 2Strength: 1
Hash
AutoV25D4C512825
Trigger Words
dark blue eyes, white pupils, black hair, blue highlights, short hair, asymmetrical hair, side braid, single braidscar on cheekmouth mask, grey earpiece, hood up, black cloak, red two-sided cloak, black shirt, turtleneck shirt, white dougi, long sleeves, wide sleeves, bridal gauntlets, blue hakama pants, armored bootsyellow katana, yellow energy sword, holding katana
34110-20250118152653.png

The Fugitive Jedi Padawan "F" from Star Wars: Visions - The Village Bride. Works best around 0.8-1.0 weight. Trained on NoobAi-1.1. Request form

Warning: As this character is trained on a NoobAi and it might not work properly on illustriousXL-0.1 based models. I recommend using a NoobAi derived models, there are some in the suggested resources to try.

Always needed:
"dark blue eyes, white pupils, black hair, blue highlights, short hair, asymmetrical hair, side braid, single braid"

  • Add "scar on cheek" when not using the Mask, note its pretty inconsistent.

Suggestions/notes:

  • Main Outfit: "mouth mask, grey earpiece, hood up, black cloak, red two-sided cloak, black shirt, turtleneck shirt, white dougi, long sleeves, wide sleeves, bridal gauntlets, blue hakama pants, armored boots"

    • Add "yellow katana, yellow energy sword, holding katana" for her weapon.

    • Remove "hood up" to make the hood go down.

    • "covered mouth" helps avoid a mouth pattern appearing on the mask.

    • Remove "mouth mask, grey earpiece" if you dont want to use the mask.

With Haru ported I moved over to bringing F up to Illust, generally works a lot nicer than Pony but the scar is still a bit funky, I considered baking longer but had worries of burning the lora so didn't push any further. I hope you all enjoy.

Feedback and reviews are always appreciated.

Training data is in the Lora metadata.

Trained on 53 images.