Wan 2.7 Pro
Wan 2.7 Pro is an image generation model from Wan that supports source image input and configurable output dimensions.
Image generation with source image support
Wan 2.7 Pro is an image generation model published by Wan and served through the Wavespeed provider on MindStudio. Released in February 2026, it is priced at $0.075 per image and accepts a context window of up to 2,000 tokens. The model supports source image input, allowing users to supply reference images as part of the generation process, and exposes controls for output width, height, seed, and a thinking mode selector.
Wan 2.7 Pro is suited for workflows that require image generation guided by an existing source image, such as style transfer, image variation, or reference-based composition. Its configurable seed input enables reproducible outputs, which is useful for iterative creative work or quality assurance pipelines. The model is available directly on MindStudio without requiring separate API key management.
What Wan 2.7 Pro supports
Source Image Input
Accepts one or more reference images as input to guide the generation process, enabling reference-based or variation-style outputs.
Custom Output Dimensions
Allows users to specify exact pixel width and height for generated images, giving precise control over output resolution.
Reproducible Seed Control
Accepts a numeric seed value so that identical inputs produce consistent, repeatable image outputs across runs.
Thinking Mode
Offers a selectable thinking mode that adjusts how the model processes prompts, providing options for different generation behaviors.
Generation Mode Toggle
Includes a toggle group input for switching between distinct generation modes, allowing users to change the model's output approach without modifying other parameters.
Ready to build with Wan 2.7 Pro?
Get Started FreeCommon questions about Wan 2.7 Pro
How much does Wan 2.7 Pro cost per image?
Wan 2.7 Pro is priced at $0.075 per generated image.
What is the context window for Wan 2.7 Pro?
The model supports a context window of up to 2,000 tokens.
Can I supply my own image as input to Wan 2.7 Pro?
Yes. The model accepts an array of image URLs as input, which are used as source or reference images during generation.
When was Wan 2.7 Pro released?
Wan 2.7 Pro was released in February 2026.
Does Wan 2.7 Pro support reproducible outputs?
Yes. The model includes a seed input, so providing the same seed and prompt combination will produce consistent results across runs.
Parameters & options
Input images to edit (1-3). Reference them as "Figure 1", "Figure 2", etc. in the prompt.
Enable thinking mode for enhanced reasoning and better image quality. Increases generation time.
Explore similar models
Start building with Wan 2.7 Pro
No API keys required. Create AI-powered workflows with Wan 2.7 Pro in minutes — free.