
LoHa of Smart Falcon (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 uploaded images are generated by AOM3A1
All of my works is free. If you'd like to support me, feel free to buy me a coffee. ?
Recommended options
-
LoRA weight around 0.6
-
Trigger words
-
Base
-
smart falcon \(umamusume\)
-
-
Racing competition costume :
-
hair bow, hair ribbon, puffy short sleeves, wrist cuffs, wrist scrunchie, collared shirt, white shirt, suspenders, black bowtie, center frills, back bow, suspender skirt, pink skirt, high-waist skirt, frilled skirt, dress, bridal garter, thigh strap, frilled socks, red socks, red footwear, sneakers
-
(optional 1) holding microphone, music, singing
-
(optional 2) stage lights, glowstick
-
-
α-star :
-
official alternate costume, hair bow, star hair ornament, choker, fingerless gloves, yellow gloves, crop top, sleeveless jacket, midriff, navel, belt, yellow bow, white skirt, pleated skirt, shorts under skirt, black shorts, bike shorts, boots
-
-
SR Guts :
-
hair bow, hair ribbon, sleeveless, strap slip, pink shirt, pink camisole, tank top, bare shoulders, sweatpants, black pants, leggings, yoga pants
-
-
Casual :
-
casual, hair bow, sleeves past wrists, puffy long sleeves, frilled collar, pink sweater, clothes writing, brown skirt, pleated skirt, plaid skirt, zettai ryouiki, miniskirt, white thighhighs, loafers, brown footwear
-
-
Tracen summer uniform :
-
tracen school uniform, summer uniform, serafuku, puffy short sleeves, purple bowtie, horseshoe ornament, sailor collar, sailor shirt, purple shirt, white skirt, pleated skirt, frills, miniskirt, zettai ryouiki, white thighhighs, brown footwear, loafers
-
-
Use a model derived from or mixed with animefull model.
-
Settings
-
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.
-
Thank you all a lot!
-
based on Animefull-pruned model
-
130 images consist of
-
4x low quality 45 images
-
8x medium quality 42 images
-
16x high quality 43 images
-
=> 1 epoch = 1204 images
-
-
replaced character feature tags with smart falcon \(umamusume\)
-
bangs, horse ears, red eyes, solo, simple background, twintails, medium hair, orange eyes, puffy sleeves, horse girl, yellow eyes, animal ears, 1girl, green eyes, hair between eyes, brown hair, horse tail, brown eyes
-
I think these tags are the essentials of Smart Falcon.
-
or unnecessary for training because of duplication.
-
-
resolution : 768x768 with aspect ratio bucketing.
-
other training settings is included in training_data.zip or metadata in LoRA file.
And, special thanks to my best friend. DH.
描述:
LoHa of Smart Falcon (umamusume). Please install extension on WebUI.
After installing extension, put LoHa file in your sd/models/lora folder and write LoRA notation in prompt to applying LoHa. <lora:loha_name:weight>
训练词语: smart falcon \(umamusume\)
名称: img.zip
大小 (KB): 504924
类型: Training Data
Pickle 扫描结果: Success
Pickle 扫描信息: No Pickle imports
病毒扫描结果: Success
名称: smart_falcon_loha-000010.safetensors
大小 (KB): 19258
类型: Model
Pickle 扫描结果: Success
Pickle 扫描信息: No Pickle imports
病毒扫描结果: Success