━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

🎯 Qwen-Image 2.1 Glamour Realism LoRA v1

Natural-looking skin and photographic colour for glamour portraits on Qwen-Image 2.1.

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

A small style LoRA (rank 16, about 80 MB) trained locally on 45 AI-generated portraits of adult women. It aims to reduce the plastic-skin and washed-out look in glamour-style portraits while keeping the outfit, pose and framing you prompt for. Use it as a gentle polish: recommended strength is 0.6.

✨ Features

• Aims for more natural skin and richer, more photographic colour in glamour portraits

• Gentle by design: at 0.5 to 0.7 it keeps outfit, pose and framing close to your prompt

• Small file (about 80 MB), no trigger word

• Tested on a 16 GB GPU

📦 Required Models

• Base model: Qwen-Image 2.1 (this LoRA is for Qwen-Image 2.1 only)

• LoRA file: qwen21_glamour_realism_v1.safetensors (about 80 MB)

• Not compatible with the original Qwen-Image (20B): LoRAs are not interchangeable between the two

• Use the same Qwen-Image 2.1 files you already use for your workflow

🧩 Required Custom Nodes

• None specific. Any standard LoRA loader node that works with Qwen-Image 2.1 is fine

🚀 How to Use

• Step 1: load Qwen-Image 2.1 as usual

• Step 2: add a LoRA loader after the model loader and pick qwen21_glamour_realism_v1.safetensors

• Step 3: set strength to 0.6

• Step 4: prompt in plain sentences describing the woman, outfit, setting and light (like the training captions). No trigger word needed

⚙️ Settings & Parameters

• Strength: 0.6 recommended, 0.5 to 0.7 usable

• At 0.8 and above, outfit and pose may start to drift from your prompt

• The strength and prompt of each example image are in its generation data

💡 Tips

• If skin looks too polished or an outfit changes, lower the strength to 0.5

• It keeps a soft, polished glamour finish. It is not a raw or gritty skin-texture LoRA

• Not tested stacked with other style LoRAs

🔗 More from me

• My Qwen-Image 2.1 Glamour workflows (SFW and After Dark) are on my profile

📝 Notes & AI Disclosure

• All example images are AI-generated with this LoRA on Qwen-Image 2.1

• Training images were AI-generated too. All subjects are adult women and none are real people

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

⭐ Found this useful?

• Like if it saved you time

• Comment your results, I read every one

• Follow for new LoRAs and ComfyUI workflows, all tested on 16 GB VRAM

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

⚖️ Model Attribution & Licensing

• Built with Qwen

• Base model: Qwen-Image 2.1 by the Qwen team. Please follow its license when you use it

• This LoRA is an adapter I trained myself and contains no Qwen model weights