Nishino Flower (umamusume)

LoCon of Nishino Flower (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>

It is highly recommend NOT using model which has high percentage of photorealistic model.

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

Recommend options

• LoRA weight 0.7 ~ 0.9

• Trigger words

  • base

    • nishino flower \(umamusume\)

  • Original art clothes :

    • red ribbon, collared shirt, striped shirt, white shirt, long sleeves, pleated dress, yellow dress, pleated skirt, belt, black footwear, shoes

  • Racing competition costume :

    • purple bowtie, striped bowtie, bare shoulders, wrist cuffs, white shirt, button, white dress, sleeveless dress, white skirt, yellow skirt, frilled skirt, thigh strap, yellow footwear, knee boots

  • Casual clothes :

    • casual, long sleeves, neck ribbon, collared shirt, blue shirt, blue dress, pink jacket

  • 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

  • VAE depends on models.

  • 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. Thank you a lot!

• based on Animefull-pruned

• 59 images of nishino flower

  • 4x low quality 19 images

  • 8x medium quality 20 images

  • 16x high quality 20 images

  • => 1 epoch = 556 images

• replaced character feature tags with nishino flower \(umamusume\)

  • horse girl, horse tail, brown hair, purple 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.

All uploaded images are generated by pastelmix(huggingface, civitai)

Thank you KohakuBlueleaf for introducing LoCon.

And, special thanks to my best friend. DH.