This LoRA was trained on a carefully curated dataset (~67 images) to capture a consistent female character with realistic facial features, natural proportions, and photorealistic skin details.

The model is optimized for SDXL photorealistic workflows and performs best with natural lighting, portrait photography, and full-body shots. It preserves facial identity well across different poses, outfits, emotions, and environments.

Recommended use:

• Photorealistic portraits and lifestyle photography

• Full body character renders

• Casual, fashion, and swimwear styles

• Mild NSFW supported depending on prompts

Training focus:

• Accurate face structure and eye color

• Natural skin texture and lighting realism

• Consistent anatomy and body proportions

• Versatility across angles and expressions

Works well with realistic checkpoints and low-to-moderate LoRA weight (≈0.6–0.9 recommended).