Wan 2.7 Image to Image: Edit and Combine Your Photos
Last verified: October 19, 2026
Wan 2.7 edits the photos you upload by following a written instruction. You add one or more images, describe the change in plain language and get back an edited picture. That makes Wan 2.7 image-to-image work a good fit for jobs you would normally give a retoucher: swapping a backdrop, restyling an outfit, placing a product in a lifestyle scene or pulling elements from several photos into one frame.
Alibaba documents two jobs for Wan 2.7 when images go in. The first is editing an existing picture from a text instruction. The second is multi-image reference generation, where several inputs feed one new composition. Both start from what is actually in your files, so the quality of your sources and how precisely you word the instruction decide most of the result.
The form keeps the workflow short. Your images go in Add Image and the instruction goes in the Prompt box. Aspect Ratio, Resolution Quality and Seed shape the output before you click Generate. The sections below cover what the model handles well, how to word edits so the right parts change, and what to do when a result misses.
Explore more image to image
Wan 2.7 Editing at a Glance
The inputs, output sizes and frame shapes this generator works with.
Input
Up to 9 images (JPG, JPEG, PNG or WEBP, up to 10 MB each) + a text prompt of up to 2,048 characters
Resolution
1K or 2K
Aspect Ratio
1:1, 3:4, 4:3, 9:16, 16:9
Output
1 image per run
Prompting Techniques for Better Wan 2.7 Edits
Naming Uploads as image 1, image 2 in the Prompt
Stating What Must Stay Unchanged
Asking for Matched Lighting When You Insert an Object
How to Edit Images with Wan 2.7
Five steps from upload to download.
Wan 2.7 Here or Through the Alibaba Cloud API
When the browser form fits the job, and when calling the model directly makes more sense.
| Criterion | Wan 2.7 | Alternatives | Best For |
|---|---|---|---|
| Getting to the first edit | Everything happens in a browser form: add images, write the instruction, choose the output settings and click Generate. | Calling Wan 2.7 through the Alibaba Cloud Model Studio API starts with creating an API key and sending requests from your own code. | Use the form for hands-on editing one image at a time, and the Model Studio API when edits need to run inside your own app or pipeline. |
| Output shape | The Aspect Ratio control offers square, portrait and landscape presets that cover common social and web formats. | The Model Studio API either keeps the shape of the last input image or accepts an exact width and height, with any ratio from 1:8 to 8:1. | Stay with the presets for standard deliverables; reach for the API when a brief calls for an exact pixel size or a very tall or very wide ratio. |
| Large source files | Each upload has to fit within the per-file size cap given in the upload FAQ below. | Model Studio accepts Wan 2.7 input images up to 20 MB each, with sides from 240 to 8,000 pixels. | Phone shots and web exports can go straight into the form; for uncompressed originals you would rather not downsize, the API's higher file limit is the better route. |
Fixing Common Wan 2.7 Editing Problems
What usually goes wrong with uploads, prompts and output sharpness, and how to fix it.
My photo won't upload in Add Image
The Prompt box won't take any more text
The edited image looks softer or smaller than my original
Frequently Asked Questions
What image files can I upload to Wan 2.7, and how many?
Add Image takes JPG, JPEG, PNG and WEBP files, 1 to 9 per run. Each file can be up to 10 MB and must be at least 300×300 pixels. You need at least one image before you can generate.
Will Wan 2.7 keep my subject's face or product the same?
Subject consistency is one of three areas where Wan 2.7 is documented with improved performance. The other two are text rendering and following complex instructions. It's a strength to lean on, not a guarantee. Name the parts that must not change in your instruction, and check faces and logos at full size before you use the result.
How is the cost of a Wan 2.7 run shown?
Once you're signed in, the credit cost of a run appears above the Generate button. It is calculated from the options you choose. Enhance Prompt suggestions don't use credits.
How many images does one Wan 2.7 run return?
Each run returns one edited image. To compare alternatives, run it again with reworded instructions or a different Seed number, and keep the version you like best.
Can I edit the result again with Wan 2.7?
Yes. Download it and add it back in Add Image as the starting image for your next instruction, as long as the file fits the upload limits above. Making one change per run makes it easier to see which instruction caused which change.