LoHa of T.M. Opera O (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. 👍

Recommended options

• LoRA weight 0.7 ~ 0.8

• Trigger words

  • Base

    • t.m. opera o \(umamusume\)

    • Mini crown and ear piercing are not included!

  • Hair ornament (not necessary for new year costume)

    • mini crown, ear piercing

  • Racing competition costume :

    • mini crown, ear piercing, white shirt, shoulder armor, pink cape, puffy sleeves, long sleeves, jewelry, fingerless gloves, single glove, white gloves, multiple rings, corset, white skirt, pink skirt, two-tone skirt, pleated skirt, white thighhighs, zettai ryouiki, high heel boots, yellow footwear

  • New year:

    • official alternate costume, japanese clothes, kimono, blue jacket, shirt, epaulettes, aiguillette, long sleeves, wide sleeves, white gloves, sash, hakama skirt, obi, floral print, high heel boots

  • Casual

    • ear piercing, white shirt, dress shirt, collared shirt, black jacket, open clothes, open jacket, long sleeves, shirt tucked in, belt, plaid pants

  • Tracen summer uniform

    • tracen school uniform, summer uniform, serafuku, puffy short sleeves, bowtie, horseshoe ornament, sailor collar, sailor shirt, frills, white skirt, purple shirt, miniskirt, zettai ryouiki, white thighhighs, brown footwear, shoes

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

• 97 images consist of

  • 4x low quality 35 images

  • 8x medium quality 30 images

  • 16x high quality 32 images

  • => 1 epoch = 892 images

• replaced character feature tags with t.m. opera o (umamusume)

  • Relation with Uma Musume

    • 1girl, solo, horse girl, horse tail, horse ears, animal ears, tail through clothes,

  • Hair

    • hair over one eye, bangs, brown hair, short hair, orange hair, multicolored hair, blonde hair

  • Eye

    • purple eyes,

  • Unnecessary

    • simple background,

• I think these tags are

  • the essentials of T.M. Opera O

  • or unnecessary for training because of duplication.

  • t.m. opera o \(umamusume\) contains above tags.

• resolution : 768x768 with aspect ratio bucketing.

• other training settings is included in training_data.zip or metadata in LoRA file.

All uploaded images are generated by AOM3A1

And, special thanks to my best friend. DH.