Sign In

Haru PonyXL (Star Wars: Visions - The Village Bride) | LoCon

69

Download

1 variant available

SafeTensor

61.11 MB

Verified:

Type

LyCORIS

Stats

164

69

31

Reviews

Published

Apr 20, 2024

Base Model

Pony

Hash

AutoV2
1C0B3ADB66

Trigger Words

tan, white pupils, aqua eyes, blonde hair
HMain, facial mark, white horns, hair flower, horn ribbon, short hair, medium breasts, white dress, off shoulder, strapless, bare shoulders, sleeveless, bare arms, white sash, vertical-striped dress, bodypaint
HaruSW, long hair, blonde hair, earrings
red robe, strapless, white shirt, blue sash, white skirt, long skirt, sleeveless
default creator card background decoration
Chameleon Badge
Manityro's Avatar

Manityro

Haru from Star Wars: Visions - The Village Bride. Works best around 0.8-1.0 weight. Training done on Pony Diffusion V6 XL. Request form

Trigger: "tan, white pupils, aqua eyes, blonde hair" always needed then pair with one of the 2 appearances below.

  • Wedding appearance:

    • "HMain, facial mark, white horns, hair flower, horn ribbon, short hair, medium breasts"

      • Outfit: "white dress, off shoulder, strapless, bare shoulders, sleeveless, bare arms, white sash, vertical-striped dress, bodypaint"

  • Casual appearance:

    • "HaruSW, long hair, blonde hair, earrings"

      • Outfit: "red robe, strapless, white shirt, blue sash, white skirt, long skirt, sleeveless"

Suggestions/notes:

  • Main Appearance notes:

    • Not very stable or flexible but sadly there was only 1 good piece of fanart to work with and the anime is rather limited for this outfit.

  • Casual Appearance notes:

    • Much more flexible and generally better looking so I would recommend it for other outfit gens.

  • Reduce weight if anything bleeds or you just wants weaker style, sadly it cannot be removed completely as the data was almost 95% from the anime episode.

Bringing up Haru from 1.5 to PonyXL and generally works better along with the main outfit being learned better than in 1.5 even with how bad the data was. I hope you all enjoy.

Feedback and reviews are always appreciated.

Training data is in the Lora metadata.

Trained on 40 images.