qwen-image-2.0
Qwen
Unifies generation and editing in one model; native 2K; professional typography and infographics.
—
¥0.08/ image
—
—
Content-rich: supports up to 4.5k token input and dense information layouts within images, generating complex layouts such as newspapers, storyboards, menus, and exam papers in a single pass. True to detail: renders small text down to 10px precisely, and faithfully reproduces micro-expressions, pores, hair strands, and other details for a texture close to real photography. Deep in knowledge: natively renders 12 languages and multiple fonts, simulates mainstream web, gaming, and livestream interfaces, and incorporates external knowledge. Qwen-Image-3.0-Pro is not just after "good-looking" but "good to use" — making image generation a truly practical productivity tool.
Entrée image 1K
/ image
Entrée image 2K
/ image
Sortie image 1K
/ image
Sortie image 2K
/ image
Entrée
Texte / Image
Sortie
Image
Facturé à l'image, environ ¥0.1 / image
Facturé par image générée avec succès, quelle que soit la résolution ou la longueur du prompt. Estimation indicative ; seul le montant réellement débité fait foi.
输出图像数量,须为整数,传字符串会返回400错误
随机数种子,未传入时由服务随机选择;固定种子可使结果相对稳定
是否开启思考模式以提升出图质量,仅在 prompt_extend=true 时生效
反向提示词,描述不希望出现的内容;仅 qwen-image-3.0 系列支持
是否开启提示词智能改写,建议开启,对简单提示词提升明显
提示词改写方式,direct 适用多数场景,agent 仅支持文生图
是否在输出图像上添加水印
