LoCon of Nakayama Festa (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. 👍

[2023-03-12 14:44 GMT] Fixed typo.

Recommend options

• LoRA weight 0.7~0.8

• Trigger words

  • base

    • nakayam festa \(umamusume\)

  • Racing competition costume :

    • grey headwear, beanie, vest, collared shirt, white shirt, long sleeves, blue jacket, letterman jacket, open clothes, black gloves, fingerless gloves, blue jacket, white skirt, pleated skirt, zettai ryouiki, black thighhighs, black footwear, sneakers

    • (another) grey headwear, bracelet, torn clothes, red shirt, striped shirt, belt, chain necklace

    • (optional) lollipop, teeth hold, grin, playing card, headphones around neck, four-leaf clover,

  • SSR Wisdom clothes:

    • official alternate costume, ahoge, halterneck, bare shoulders, jewelry, necklace, bead bracelet, swimsuit, navel, midriff, black bikini, black shorts, bikini shorts, sandals

    • (optional) holding water gun,

  • Casual clothes :

    • casual, beanie, long sleeves, white hoodie, denim jacket, blue jacket, open clothes, open jacket, short shorts, black shorts

  • Tracen school uniform :

    • tracen school uniform, purple shirt, pleated skirt, puffy short sleeves, white skirt, puffy sleeves, summer uniform, frilled skirt, sailor collar, sailor shirt, miniskirt, frills, white thighhighs

Use animefull based model

• Settings

  • Stable Diffusion WebUI by Automatic1111

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

  • about 20 steps, CFG scale 3.5~6.5

  • I recommend kl-f8-anime2.vae

  • CLIP skip = 1

  • Use Hires.fix to get higher quality image

    • Up-scaler : Latent series

    • Denoising strength 0.50~0.65

LoRA training info

• trained on sd-scripts by kohya_ss.

  • LyCORIS by KohakuBlueleaf

  • Thank you all a lot!

• based on Animefull-pruned model

• 66 images of el condor pasa

  • 4x low quality 29 images

  • 8x medium quality 12 images

  • 16x high quality 25 images

  • => 1 epoch = 612 images

• replaced character feature tags with nakayama festa \(umamusume\)

  • horse girl, horse tail, brown hair, red eyes, etc.

• resolution : 768x768

• rank=alpha=16

• conv_rank=conv_alpha=16

• batch size = 4

• other training settings is included in training_data.zip

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

All uploaded images are generated by AOM3A1

Settings of all preview images are same, only LoRA differs.

And, special thanks to my best friend. DH.