All the models in one place
Run GPT Image 2.6 and every other model on Glif from one chat. One account, one balance, no API keys to wire up.
Retired: Glif no longer uses this model. Try GPT Image 2.6 instead.
Qwen Image Generator with LoRA
Qwen Image with LoRA is Alibaba’s 20-billion-parameter image model with support for custom adapters, now retired on Glif.
Test outputs
Compare all modelsQwen Image Generator with LoRA in practice
What it does and when to use it.
Model summary
Qwen Image with LoRA is Alibaba’s 20-billion-parameter image model with support for custom adapters, now retired on Glif. It generates from text or edits an input image at 512 to 2048 pixels per side, with up to 3 LoRA adapters for style and subject. It renders legible text and suits stylized illustrations and character LoRAs. Skin and lighting often look airbrushed. Glif charges per image.
Capabilities
Text to image
Portrait generation
Selfie style images
In image text rendering
Scene composition
Lighting and reflections
Atmospheric effects
Object ordering and layout
Lora customization
Semi photorealistic illustration
Suggested use cases
- Stylized illustrations with a custom LoRA
- A recurring character from a trained LoRA
- Images with short, legible text