LoRA of Mayano Topgun (umamusume)

NOT COMPATIBLE WITH AOM3A2, AOM3A3 model.

You Copy? I Copy~

Recommend options

• LoRA weight 0.6~0.9

• Trigger words

  • base

    • mayano topgun \(umamusume\)

  • Racing competition clothes :

    • dog tags, navel, brown thighhighs, midriff, yellow crop top, bomber jacket, white short shorts, green jacket, boots, brown footwear, open jacket

    • (optional) : thumbs up, sunglasses, aircarft

  • Casual clothes :

    • long sleeves, white shirt, plaid skirt, pleated skirt, yellow skirt, white belt, belt buckle, yellow vest, orange skirt

  • Wedding costume :

    • hair ornament, hair flower, official alternate costume, pantyhose, strapless dress, white gloves, fishnets, wedding dress, bare shoulders, pink dress, choker, holding bouquet, shoes, jewelry

  • Christmas costume :

    • christmas, red dress, red capelet, belt, santa hat, fur-trimmed capelet, santa costume, red headwear, official alternate costume, bell, ears through headwear, fur-trimmed dress

Use animefull based model

• Settings

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

  • about 20 steps, CFG scale 3.5~6.5

  • kl-f8-anime2.vae

  • CLIP skip = 1

  • If using highres

    • 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

• 206 images of mayano topgun

  • 4x low quality 78 images

  • 8x medium quality 68 images

  • 16x high quality 60 images

  • => 1 epoch = 1816 images

• replaced character feature tags with mayano topgun \(umamusume\)

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

• using aspect ratio bucketing

• resolution : 768x768

• rank=alpha=16

• using dadaptation

• batch size = 4, 8 epochs trained

• clip skip = 1

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

This LoRA can generate nsfw images of mayano topgun.

All uploaded images are generated by AOM3A1