QwenLM/Qwen-Image — reverse-engineered prompt

Build me a Python project for Qwen Image, an open source image generation and editing model.

I want it to feel like a simple demo I can run locally. I should be able to type a prompt and generate an image, especially prompts that include visible text on the image. It should also let me upload an image and describe an edit, then save the edited result. Please make the interface friendly, with clear buttons, example prompts, progress messages, and a results area where I can download images.

Use the public Qwen Image models from Hugging Face or ModelScope, and make setup as painless as possible. Include a clear README with install steps, hardware notes, how to choose text to image versus editing, and a few examples in English and Chinese. If model files are too large, don’t bundle them, just download or load them properly.

Look up the current docs online if you need to.