Wan 2.7
Wan 2.7 is an image generation model from Alibaba's Wan that supports source image inputs and custom dimensions at $0.03 per image.
Image generation with source image support
Wan 2.7 (wan2.7-image) is an image generation model published by Wan and provided through Alibaba, released in February 2026. It accepts source images as input alongside configurable width and height parameters, making it suited for image-to-image workflows rather than purely text-to-image generation. A seed input is also supported, allowing users to reproduce specific outputs consistently across runs.
The model operates with a 2,000-token context window and is priced at $0.03 per generated image. Its input structure — including a mode toggle, source image array, and numeric dimension controls — indicates it is designed for workflows that require transformation or conditioning on existing visual content. It is available on MindStudio without requiring separate API key configuration.
What Wan 2.7 supports
Source Image Input
Accepts one or more source images as input via an image URL array, enabling image-to-image generation workflows rather than text-only prompting.
Custom Dimensions
Allows users to specify exact output width and height as numeric values, giving direct control over the generated image's resolution.
Seed Control
Supports a seed parameter so that specific generation results can be reproduced exactly by reusing the same seed value.
Mode Selection
Includes a toggle group input for switching between generation modes, allowing different processing behaviors within the same model.
Ready to build with Wan 2.7?
Get Started FreeCommon questions about Wan 2.7
How much does Wan 2.7 cost to use?
Wan 2.7 is priced at $0.03 per generated image.
What is the context window for Wan 2.7?
Wan 2.7 has a context window of 2,000 tokens.
Does Wan 2.7 support image inputs?
Yes. Wan 2.7 accepts source images via an image URL array input, making it suitable for image-to-image generation tasks.
Can I control the output image size?
Yes. The model accepts numeric width and height inputs, allowing you to specify the exact dimensions of the generated image.
When was Wan 2.7 released?
Wan 2.7 was released in February 2026.
Documentation & links
Parameters & options
Input image(s) to edit or use as references. Supports multi-image reference generation — reference them in your prompt.
Explore similar models
Start building with Wan 2.7
No API keys required. Create AI-powered workflows with Wan 2.7 in minutes — free.