About This Model
Wanxiang 2.7 (万相 2.7) is the latest image generation model from Alibaba's Wanxiang family, available via Alibaba Cloud Model Studio. It offers Pro and Standard variants — Pro supports 4K output and thinking mode for higher quality, while Standard provides faster generation at lower cost.
Key Features
- Pro & Standard versions — Pro for best quality with 4K and thinking mode, Standard for fast generation
- Thinking mode — AI-enhanced reasoning for better prompt understanding and image quality (Pro text-to-image)
- Up to 4K resolution — Pro supports 1K, 2K, and 4K output
- Image editing — Edit existing images with text instructions, support multi-image input (up to 9 images)
- Multi-image generation — Generate up to 4 images per request for rapid iteration
- Diverse art styles — Strong in photorealism, illustration, and various artistic styles
Best For
- High-resolution (4K) image generation for print and commercial use
- Image editing: object manipulation, style transfer, background replacement
- Rapid creative iteration with multi-image output
- Photorealistic scenes and diverse artistic styles
FAQ
What's the difference between Pro and Standard?
Pro (wan2.7-image-pro) supports 4K output, thinking mode, and produces higher quality results. Standard (wan2.7-image) is limited to 2K but generates faster and costs less — about 40% of Pro's price.
What is thinking mode?
When enabled (default), the model uses enhanced reasoning to better understand complex prompts, improving output quality at the cost of slightly longer generation time. Only available for Pro text-to-image.