Downloads · 30 days
15
7% of all-time downloads
JunkieMonkey69/HunterSchafer_ZimageTurbo
HunterSchafer_ZimageTurbo is a text-to-image model from JunkieMonkey69. Use it when you need an image from a text prompt. It is set up for diffusers. The card lists the license as apache-2.0.
Downloads · 30 days
15
7% of all-time downloads
All-time downloads
221
Public
Repo size
174 MB
Likes
0
Public
Click a slice to open those files.
.safetensors170 MB · 98%
From the Hugging Face model README
🧩 LoRA Name: HunterSchafer_Actress
>Base Model: Tongyi-MAI / Z-Image-Turbo >LoRA Type: Character / Actress >Trigger Word(s): Hunter Schafer >Recommended Weight: 0.75 – 1 (sweet spot ≈ 0.85)
✨ Overview
>This LoRA captures the distinctive facial structure, styling range, and editorial presence of Hunter Schafer, optimized for ZimageTurbo. It is trained to preserve identity consistency across varied fashion, lighting, and mood setups, performing well in both clean studio portraits and stylized editorial scenes.
The model balances realism with high-fashion expressiveness, making it suitable for portrait photography, magazine-style shoots, and cinematic close-ups.
🎨 What This LoRA Is Good At
>Strong facial identity retention across angles and lighting >Editorial and fashion-oriented aesthetics >Clean skin rendering with natural facial detail >Works well with both minimal and stylized makeup >Consistent results across portrait and mid-shot compositions
⚙️ Usage Notes
>Best results with portrait, upper-body, and editorial framing >Prompting the name “Hunter Schafer” is sufficient; no extra trigger token required >Avoid stacking multiple strong character LoRAs simultaneously >Performs especially well with fashion, studio, and cinematic lighting prompts
🧪 Example Prompt Hunter Schafer, high-fashion editorial portrait, soft studio lighting, clean background, cinematic composition, detailed facial features
📐 Training Details
>Dataset Size: 46 images (fully captioned) >Training Resolutions: 512 / 768 / 1024 >Total Steps: 4,600 >Batch Size: 1 >Content vs Style: Balanced >Noise Scheduler: FlowMatch >Optimizer: AdamW 8-bit >Learning Rate: 1e-4 >Precision: BF16 >Text Encoder Training: Disabled >UNet Training: Enabled >Quantization: QFloat8 (model + text encoder) >Architecture: ZImage Turbo LoRA (LoKR full-rank enabled)
⚠️ Limitations / Notes
>Not designed for anime or heavily stylized cartoon outputs >Extreme fisheye angles or exaggerated facial distortion may reduce likeness >Best results achieved with guidance scale ≤ 1 (ZimageTurbo native behavior)
You should use Hunter Schafer to trigger the image generation.
Download them in the Files & versions tab.