Wan-2.7 Image-to-image
Edits and recomposes images with Wan 2.7 image using text instructions, multi-image references, and optional interaction boxes.
What it does
Wan-2.7 Image-to-image, in practice.
Alibaba WAN 2.7 Image Edit enables instruction-driven image editing and multi-image composition. It is designed for workflows where you want to preserve the source scene, combine references, and steer the result with natural language.
- Instruction-based editing: Rewrite clothing, materials, color, scene details, or object relationships with a plain-language prompt.
- Single-image and multi-image input: Works well for focused edits as well as reference-driven composition.
- Reference composition: Combine multiple images and a text instruction in one edit request.
- Practical controls: Supports `size`, `n`, `watermark`, `thinking_mode`, and `seed`.
- Marketing teams adapting hero visuals for multiple campaigns.
- E-commerce teams refreshing product shots and styled composites.
Run Wan-2.7 Image-to-image
from $0.045/imageParameters
What you can set.
| Parameter | What it does | Options |
|---|---|---|
prompt | Text instruction used to edit the image. Maximum length is 5000 characters. | |
images | Input images, each provided as a URL or Base64 string. The first image is treated as the primary input; subsequent images serve as references. | |
size | Output image resolution. | 1K 2K |
n | Number of images to generate. | default 1 |
thinking_mode | Whether to enable thinking mode for higher-quality image generation. | default True |
seed | Random seed for image generation. Range: 0 to 2147483647. Use -1 for a random seed. | default -1 |
Sample prompt
The prompt behind the sample.
Adjust the characters to look like real people with realistic facial features, natural skin texture, and human proportions.
size: 2Kn: 1FAQ
Short answers.
How much does Wan-2.7 Image-to-image cost?
Pricing starts at $0.045 per image at the base resolution; higher resolutions and longer durations cost more. Usage is billed per request from your balance — no subscription.
Does Wan-2.7 Image-to-image run uncensored here?
No. Qwen applies its own content policy to this model regardless of where it is called from. For a route without an extra platform filter use Uncensored MiniMax H3; this page lists Wan-2.7 Image-to-image for everything else it does well.
What does Wan-2.7 Image-to-image take as input?
It is a image editing model. Edits and recomposes images with Wan 2.7 image using text instructions, multi-image references, and optional interaction boxes.
How do I use it?
Enter an invite code to open Chat with the model loaded, or join the waitlist. We open seats in batches and track which models are requested most.
Related