LoHa of Darley Arabian (umamusume). Put LoHa file in your stable-diffusion-webui/models/lora folder and write LoRA notation in prompt to applying LoHa. <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. 👍

NOTICE

• This is the last LoRA of each character of Umamusume : Pretty Derby.

• I want to thank everyone. The process of producing these was a very interesting experience. It would not have been possible without you. Again, thank you.

• And, special thanks to dear my best friend. DH.

Recommended options

• LoRA weight 0.7 ~ 0.8

• Trigger words

  • Base

    • darley arabian \(umamusume\)

  • Original Art :

    • ear ornament, jewelry, necklace, bra strap, crop top, horseshoe ornament, frilled shirt, red shirt, bare shoulders, off-shoulder shirt, detached sleeves, puffy long sleeves, frilled sleeves, bracelet, highleg panties, jeans, blue pants, sandals, white footwear

Use a model derived from or mixed with animefull 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

    • Upscaler

      • Latent, Latent (nearest-exact), Latent (bicubic antialiased) or other 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

• replaced character feature tags with darley arabian \(umamusume\)

  • 1girl, horse girl, horse tail, bangs, multicolored hair, solo, animal ears, dark skin, medium hair, transparent background, alternate costume, brown hair, simple background, horse ears, white hair, streaked hair, hair between eyes, long hair, dark-skinned female, ahoge, aqua eyes, two-tone hair, personification, red hair, pink hair, green eyes

  • I think these tags are the essentials of Darley Arabian or

  • unnecessary for training because of duplication.

• resolution : 768x768 with aspect ratio bucketing.

• other training settings is included in training data or metadata of LoRA file.

All uploaded images are generated by AOM3A1