Qwen-Image-2.1-PE-T2I is a text-to-image prompt rewriting model. It takes brief image requests in any language and generates detailed English prompts with recommended aspect ratios.
The model is a fine-tuned Qwen3.5-VL 9B. It outputs JSON containing rewritten_prompt and wh_ratio for image generation.
It supports multiple languages and integrates with Qwen-Image-2.1 for text-to-image creation.