BellaMax ZIT (Z-Image Turbo)

BellaMax ZIT is a high-fidelity likeness LoRA specifically built for the Z-Image Turbo ecosystem. While it shares the same aesthetic DNA as the Pony BellaMax, this is a ground-up training optimized for ultra-fast, low-step generation without sacrificing facial precision.

This model is designed to "lock in" the subject’s unique facial structure while remaining highly adaptable to different environments, outfits, and even dramatic hair color changes.

What’s inside the training?

Based on the high-resolution dataset, this LoRA has a strong understanding of:

• Signature Features: Small nose ring (left nostril), silver septum piercing, and a light dusting of freckles across the cheeks and nose. 

• Versatile Hair: Highly responsive to hair styling, from her signature long honey-blonde waves to vibrant red bobs with bangs or high ponytails. 

• Aesthetic Range: Native support for various "vibes" including luxury poolside settings, gritty urban alleys, cozy bedroom intimacy, and modern gym environments. 

How to Prompt for BellaMax

To get the most accurate likeness, use these descriptive tokens:

• The Face: 1girl, small nose ring, silver septum ring, light freckles, almond-shaped eyes, subtle smile.

• The Hair: * Classic: long wavy honey-blonde hair. 

• Alternative: vibrant red hair, bob cut, bangs. 

• Styles: This LoRA is excellent for specific photography styles:

• Monochrome: Use black and white, film grain, vintage filter for a timeless, moody portrait. 

• Cyber/Moody: Use purple lighting, cool tones, neon reflections for an artistic, introspective look. 

Recommended Training/Usage Specs:

• Model Type: Z-Image Turbo (ZIT)

• Steps: 4–10 steps (Turbo optimized)

• CFG: 1.0 – 2.5

• LoRA Weight: 0.8 – 1.0 (Reduce to 0.7 if the style becomes too "stiff")