Sign In

Curren Chan (umamusume)

722
7.8k
31
Verified:
SafeTensor
Type
LoRA
Stats
7,814
Reviews
Published
Feb 19, 2023
Base Model
SD 1.5
Training
Steps: 2,790
Epochs: 9
Trigger Words
curren chan \(umamusume\)
Training Images
Download
Hash
AutoV2
40DB865CC0
default creator card background decoration
Donator Badge
mht's Avatar
mht

LoRA of Curren Chan (umamusume). Put LoRA file in your stable-diffusion-webui/models/lora folder and write LoRA notation in prompt to applying LoRA. <lora:your_loha_name:weight>

All of my works is free. If you'd like to support me, feel free to buy me a coffee. 👍

Recommend options

  • LoRA weight 0.6~1.0

  • Trigger words

    • base

      • curren chan \(umamusume\)

      • (optional)smile, hands on own chin,

    • Racing competition clothes :

      • vertical-striped dress, bare shoulders, off shoulder, closed mouth, long sleeves, collared dress, gloves

    • Wedding costume :

      • elbow gloves, black wedding dress, black gloves, string of fate, official alternate costume, closed mouth, bare shoulders, sleeveless dress

    • Casual clothes :

      • jewelry, white shirt, white long sleeves, necklace, red skirt, sleeves past wrists, black pantyhose

        • maybe white shirt need more weight

  • Use animefull based model

  • Settings

    • DPM++ Series(SDE Karras, 2M Karras, etc.)

    • about 20 steps, CFG scale 3.5~6.5

    • kl-f8-anime2.vae

    • CLIP skip = 1

    • If using highres

      • Latent series

      • Denoising strength 0.50~0.65

LoRA training info

  • based on Animefull-pruned

  • 133 images of curren chan

    • 4x low quality 42 images

    • 8x medium quality 48 images

    • 16x high quality 43 images

    • => 1 epoch = 1240 images

  • replaced character feature tags with curren chan \(umamusume\)

    • horse girl, horse tail, pink hair, purple eyes, etc.

  • using aspect ratio bucketing

  • resolution : 768x768

  • rank=alpha=16

  • using dadaptation

  • batch size = 4, 9 epochs trained

  • clip skip = 1

This LoRA may be compatible with animefull-based models (ex. AbyssOrangeMix Series...)

All uploaded images are generated by AOM3A1