Hishi Amazon (umamusume)

LoRA of Hishi Amazon (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>

Old racing clothes is so hot 👍

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

Recommend options

• LoRA weight 0.7~1.0

• Trigger words

  • base

    • hishi amazon \(umamusume\)

  • Racing competition clothes :

    • crop top, bandeau, white choker, single sleeves, arm ribbon, pleated skirt, thigh strap, white skirt, , skirt, layered skirt, thigh strap, leg ribbon, blue footwear

    • (optional) hand on hip, hand on own thigh, leaning forward

  • Old racing competition clothes :

    • sailor collar, wide sleeves, blue neckerchief, white shirt, sailor dress, thigh strap, blue footwear, tanlines, tan, off shoulder, , shoulders

    • (optional) under clothes, blue skirt

  • Casual clothes :

    • red shirt, fang, torn clothes, choker, bracelet, jewelry, long sleeves, shoulders, midriff, torn jeans, torn pants, denim, red footwear, shoes, sneakers

  • Tracen school uniform :

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

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

  • kl-f8-anime2.vae

  • CLIP skip = 1

  • Use highres. fix to get higher quality image

    • Upscaler : 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

• 115 images of taiki shuttle

  • 4x low quality 39 images

  • 8x medium quality 32 images

  • 16x high quality 44 images

  • => 1 epoch = 1116 images

• replaced character feature tags with hishi amazon \(umamusume\)

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

• resolution : 768x768

• rank=alpha=16

• batch size = 4, 7 epochs trained

• other training settings included in training_data.zip

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

All uploaded images are generated by AOM3A1