[Thorne🎸| Fortnite] - 🎨Illustrious XL/🦄Pony XL LoRA

This is a high-fidelity LoRA designed for Illustrious-XL/Pony XL It was trained on a curated synthetic dataset to ensure maximum flexibility and character consistency while maintaining the high-aesthetic standards of the base model.

🎸Model Overview: Thorne

The Thorne model focuses on a high-fidelity, stylized character design that bridges the gap between Fortnite’s semi-realistic 3D aesthetic and a modern digital illustration style. It excels at capturing specific fashion textures-like wool plaid and sheer hosiery-alongside intricate skin details.

Core Visual Elements

• Hair Style: Features a distinct "two-tone" or sunset gradient, typically transitioning from a deep chestnut or mahogany at the roots to a vibrant ginger or orange at the tips.

• Wardrobe: * Top: A crisp, white short-sleeved button-down shirt.

  • Accessories: Black clip-on suspenders.

  • Bottom: A high-waisted red and black tartan (plaid) pleated mini skirt.

  • Footwear: Classic red high-top canvas sneakers with white laces.

• Body Art: Heavy emphasis on fine-line black ink tattoos, specifically floral and vine motifs covering the left arm (sleeve style) and extending onto the hand.

🚀 Usage Recommendations

⚙️Base Model: 🎨Illustrious

• Trigger Word: ThorneFN

• LoRA Weight: 0.7 - 0.8 (Start at 0.7 for maximum accuracy)

• CFG Scale: 5.0 - 7.0

• Clip Skip: 2

• Sampling Method: Euler a or DPM++ 2M Karras

• Recommend to use: Nova 3DCG XL, WAI-illustrious-SDXL

⚙️ Base Model: 🦄Pony XL

• Trigger Word: ThorneFN

• Recommended Weight: 0.8 - 0.9

• CFG Scale: 5.0 - 7.0

• Sampler: DPM++ 2M Karras (25–30 steps)

• Recommend to use: WAI-ANI-PONYXL, WAI-SemiReal, BlenderMix

🛠️ Prompting Guide

This model is highly responsive to descriptive tags. For the best results, use a mix of the trigger word and specific outfit/pose descriptions.

• Main Trigger: ThorneFN

• Core Look: 1girl, solo, orange hair, brown hair, long hair, multicolored hair, gradient hair, brown eyes, lips

• Signature Outfit: white shirt, short sleeves, collared shirt, suspenders, suspender skirt, red skirt, plaid skirt, pleated skirt, black pantyhose

• Fashion & Details: sneakers, red footwear, arm tattoo, tattoo, jewelry, fashion

• Negative Tags: low quality, blurry, deformed, text, watermark

📈 Technical Details

• Dataset: Created using a high-quality synthetic workflow to generate diverse angles and lighting from a single reference source.

• Training: Optimized using bfloat16 for precision

• Train Model: WAI-illustrious-SDXL v.17/ WAI-ANI-PONYXL v.14

• Version: 2.0/1.0 (Last Release)

🕒 Version History

Version 1.0 (Initial Release)

• Focus: Establishing character consistency.

• Note: Was 1st version of that lora (beta).

Version 2.0 (Dataset Expansion)

• Focus: Flexibility and Pose Variety.

• Details: Expanded the dataset with 20 additional synthetic images featuring different camera angles (overhead, from behind) and lighting conditions.

• Improvements: Better Quality, More Poses

Version 1.0 (Pony)

• Focus: Pony release

• Note: Establishing character consistency.

💬 Notes & Feedback

I am constantly refining the dataset generation process. If you enjoy the model, please post your generations and leave a review-it helps me understand how to improve future versions!

If you run into issues with character bleeding or specific clothing items, let me know in the discussion.