Input your prompt to generate Qwen Image
Have a different question and can’t find the answer you’re looking for? Reach out to our support team by sending us an email and we’ll get back to you as soon as we can.
Qwen Image is a 20B MMDiT image foundation model that excels at both image generation and precise image editing. It supports complex text rendering (including multi-line, paragraph-level, and bilingual text), style transfer, object addition/removal, background changes, and fine-grained detail enhancement. Qwen Image can generate photorealistic, anime, artistic, and infographic images, and is especially strong at rendering text in both English and Chinese.
For best results, use clear and detailed prompts. Specify the desired content, style, and any text to appear in the image. For text rendering, include the exact wording and placement. For editing, describe what to add, remove, or modify. Qwen Image supports both English and Chinese prompts, and can handle complex layouts and multi-element scenes.
Qwen Image supports:
Qwen Image is available for research and non-commercial use. You can experience the latest model through Flux.1 AI. Each registered user will receive 10 credits, giving you a free opportunity to try out Qwen Image.
Qwen Image achieves state-of-the-art results on multiple public benchmarks for both image generation (GenEval, DPG, OneIG-Bench) and image editing (GEdit, ImgEdit, GSO). It is especially strong in text rendering, outperforming other models on LongText-Bench, ChineseWord, and TextCraft, and is a leading model for both general and text-centric image tasks.
Qwen Image can be used for:
Qwen Image takes approximately 5 seconds to generate an image. There may be fluctuations during peak times, but it should return within 30 seconds.