Structured visual design
Create infographics, campaign layouts, and information-rich visuals with clearer hierarchy.
Generate polished 1K or 2K visuals from text, or transform reference images with stronger composition, readable text, realistic materials, and multilingual understanding.

Choose a sample to preview the model's visual range. Your generated image will replace it here.
Qwen Image 3.0 is most useful when composition, text, realism, and language all need to work together.
Create infographics, campaign layouts, and information-rich visuals with clearer hierarchy.
Generate portraits, products, and lifestyle scenes with more convincing light and material texture.
Write prompts in widely used languages and create localized visuals with stronger text understanding.
It is Alibaba Qwen's professional multimodal image model for text-to-image creation, reference-based editing, realistic detail, visual text, and multilingual generation.
All four models support 1K and 2K output. Pro costs 10 credits at 1K and 20 at 2K; Standard costs 8 credits at either resolution.
Yes. It is designed to handle posters, labels, information layouts, and other text-aware visual tasks more reliably.
Images are generated and downloaded as high-quality PNG files by default.
Yes. Both image-to-image models accept up to 3 JPG, PNG, or WebP references. Pro includes the first reference and adds 1 credit for each additional image.
Describe the subject, camera or composition, lighting, material, visual style, and required text in a clear order.