Base Style & Quality Enhancement LoRA (Pony V6)
This is a base style / image quality enhancement LoRA designed to unify artistic style, improve line clarity, and enhance overall visual depth and lighting.
It focuses purely on art style and texture, allowing you to freely change character designs, outfits, poses, and expressions.
Recommended Settings
- Base Model: Pony Diffusion V6 (or Pony-based fine-tuned models)
- LoRA Weight:
0.3 ~ 0.6 (Recommended for stacking with character or pose LoRAs)
- Clip Skip:
1
- CFG Scale:
5.0 ~ 7.0
Trigger Word
t2y_base
Include this trigger word in your prompt to apply the base style.
Example Prompt
score_9, score_8_up, score_7_up, source_anime, t2y_base, 1girl, solo, standing, full body, looking at viewer, school uniform, black hair, long hair, light smile, simple background
Key Features
- High Compatibility: Works seamlessly alongside other character LoRAs or pose LoRAs without breaking their core features.
- Flexible Concept Separation: Fully tagged dataset ensures outfits, backgrounds, and character traits remain completely customizable.
