Alf LoRA (Silver Palace)
Training Information
- Dataset: 43 Images
- Epochs: 10
- Total Steps: 1935
- Batch Size: 2
- Repeats: 9
Recommended Model
- Epoch 7 (Recommended)
Alf (フウ, 阿芙, 알프) - Silver Palace
⚙️ Recommended Settings
Weight: 0.8 ~ 1.0
Trigger Words:
alf_silverpalaceClip 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),
red cable tail, chest strap, inner strap, white strap, halterneck, bare shoulders, sideboob, white apron, white strap, detached sleeves, frills, red bowtie, medal, flower ornament, belt, pocket watch, hanging straps, black straps, black thigh strap, brass rings, white pantyhose, black mary janes,
alf_silverpalace, 1girl, solo, pink hair, short hair, ahoge, blue eyes, slit pupils, white pupils, bangs, hair ribbon, small breasts, round face,
<lora:Alf_v2:0.8>🚫 Negative Prompt Example
(worst quality, low quality:1.3), (lowres:1.2), (3d:1.1), bad anatomy, (bad hands:1.2), text, error, missing fingers, extra digit, fewer digits, cropped, jpeg artifacts, signature, watermark, username,Description
Adding Data & Changing Training Method
Reduced texture details on garments to prioritize facial details.







