成田白仁 Narita Brian

https://s2.xiaonaofu.top/material/39075692-f6c6-48b9-b99f-187af9249c22.jpeghttps://s2.xiaonaofu.top/material/8c602053-baa6-4e2d-b31a-b2c0b16d6028.jpeghttps://s2.xiaonaofu.top/material/44a9b72f-4cc8-4057-ad3c-0e1239e48598.jpeghttps://s2.xiaonaofu.top/material/8bb77e7a-5a7a-45ec-95e4-6e25cbadbb74.jpeghttps://s2.xiaonaofu.top/material/98cabdbb-a35b-4660-afba-18079217d069.jpegLoRA of Narita Brian (umamusume)

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~

Trigger words

base

narita brian \(umamusume\)

Racing competition clothes :

midriff, fingerless gloves, coat on shoulders, plaid skirt, belt, asymmetrical footwear, cleavage, single thighhigh, chest sarashi

Casual clothes :

long sleeves, white shirt, purple pants

Blaze costume NOT included.

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

based on Animefull-pruned

68 images of narita brian

4x low quality 29 images

8x medium quality 17 images

16x high quality 30 images

=> 1 epoch = 812 images

replaced character feature tags with narita brian \(umamusume\)

horse girl, horse tail, brown hair, yellow eyes, etc.

using aspect ratio bucketing

resolution : 768×768

rank=alpha=16

using dadaptation

batch size = 4, 9 epochs trained

clip skip = 1

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

All uploaded images are generated by AOM3A1

发表回复

后才能评论