Sayuri LoRA (Blue Archive)

Training Information
- Dataset: 20 Images
- Epochs: 10
- Total Steps: 2000
- Batch Size: 2
- Repeats: 20

Recommended Model
- Epoch 7 (Recommended)

"Please use the V2 LoRA!"

Sayuri (サユリ) - Blue Archive

⚙️ Recommended Settings

• Weight: 0.8 ~ 1.0

• Trigger Words: sr

• Clip Skip: 2

• Sampling method: Euler a

• Steps: 28 ~ 30

• Width, Height: 1024,1024

• CFG Scale: 5 ~ 7

🔍 Recommended Hires. fix Settings

• Upscaler: R-ESRGAN 4x+ Anime6

• Hires steps: 10

• Denoising strength: 0.4

• Upscale by: 2

Recommended Settings

- ADetailer: ON

- Upscaling: 1.5x ~ 2x

💡 Prompt Example (Base Outfit)

(masterpiece:1.1), pastel colors,
white one-piece swimsuit, highleg, detached collar, blue ribbon, white armband, white wristwatch, thigh holster, white holster,
sr, 1girl, solo, white hair, long hair, white eyelashes, blue eyes, blue halo, orca hair ornament, cross hairclip,  
<lora:Sayuri_v2:0.8>

🚫 Negative Prompt Example

(worst quality, low quality:1.2), (lowres:1.2), bad anatomy, (bad hands:1.2), (cast shadow, hard shadows:1.2), directional lighting, chiaroscurotext, error, missing fingers, extra digit, fewer digits, cropped, jpeg artifacts, signature, watermark, username