Powered by Qwen-Image-2512, Alibaba's open-source image generation model.
Experience superior human realism, native Chinese text rendering, and blazing-fast generation.
Open source under Apache 2.0 license

Qwen-Image-2512 is Alibaba's latest AI image generation model, built on a 20 billion parameter Multimodal Diffusion Transformer architecture.
Reduced AI look with accurate facial features, natural postures, and semantic-aware backgrounds.
Accurately render Chinese, English, and CJK text directly in generated images with proper layout.
Generate high-quality images in under 3 seconds, compared to 30-60 seconds for other models.
Released under Apache 2.0 license. Free to use, modify, fine-tune, and deploy commercially.
The strongest open-source image generation model with production-ready quality.
Everything you need for professional AI image generation.
Create stunning images from text prompts with exceptional detail and creativity.
Object detection, semantic segmentation, depth estimation, and edge detection.
Upscale and enhance image quality with AI-powered super-resolution.
Apply artistic styles to images while preserving content and structure.
Superior rendering of water, landscapes, fur, wood, textiles, and other natural materials.
Generate images from prompts in Chinese, English, and other languages.
Learn more about Qwen-Image-2512 and how to get started.
Experience the power of open-source AI image generation.