Eye Concept SDXL / Pony LoRA | Hyper-Realistic Iris Detail

The Definitive Eye Detail LoRA for SDXL & Pony Models

The Eye Concept LoRA specializes in adding ultra-high-resolution, photorealistic detail to the human iris, transforming standard SDXL and Pony eyes into hyper-detailed focal points. This model was trained on a consolidated dataset to provide consistent, stunning quality across multiple eye colors in a single LoRA file.

Key Features

• Universal SDXL Compatibility: Works perfectly with SDXL 1.0 and all popular Pony (e.g., Ponydiffusion V6) checkpoints.

• Consolidated Colors: Trained across multiple real-world eye colors (blue, brown, green, etc.), allowing you to prompt for any eye color while maintaining the same high-detail texture.

• Hyper-Detail Focus: Utilizes an extremely low Network Rank (Dim: 8) to dedicate its learning capacity almost exclusively to iris texture, light reflection, and internal fiber structure.

• Seamless Integration: Maintains the aesthetic integrity of the base model, ensuring perfect blending with the realism of SDXL or the stylized output of Pony.

Usage Instructions

ParameterRecommended ValueBase ModelStable Diffusion XL (SDXL) 1.0 or any Pony checkpoint
LoRA Weight0.5 to 0.7 (Start at 0.6)
Trigger Wordeye_concept (Use this in your positive prompt)Color Prompt
Add the desired color (e.g., perfect blue eyes, detailed hazel eyes)

Example Prompt Snippet:

photorealistic portrait of a woman, <lora:eye_concept:0.6>, cinematic lighting, highly detailed emerald green eyes, sharp iris texture

Notes

This is a highly specific detail LoRA. If you are using a Pony model, the base aesthetic will be retained (anime style), but the eyes will receive a massive boost in detail and realism. Adjust the weight lower (down to 0.5) if you want the base style to dominate more.