WAN

Wan 2.7 Pro Edit is a powerful image editing model that lets you transform and modify your existing images using simple text instructions. Rather than generating images from scratch, this model takes one or more reference images you provide and applies precise, professional-grade edits based on your written descriptions — making it an intuitive and versatile tool for anyone who wants to reshape, restyle, or reimagine their visual work.

At its core, the model works through a straightforward creative workflow: you upload your reference images (up to four at a time), write a text prompt describing what you want changed, and the model produces a new image that reflects your instructions. This text-driven approach to image editing means you don't need advanced software skills or hours of manual retouching — you simply describe the transformation you have in mind, and the model handles the rest.

One of the standout features of Wan 2.7 Pro Edit is its support for multi-image referencing. You can upload between one and four reference images and refer to them by number in your prompt — for example, "Turn image 1 into a watercolor painting" or "Combine the background of image 2 with the subject of image 1." The order in which you upload your images matters, as the model uses your numbering (image 1, image 2, image 3, image 4) to understand which source material you're referring to. This opens up exciting possibilities for compositing, style blending, and complex multi-source edits that would otherwise require significant manual effort in traditional editing software.

The model supports prompts in both Chinese and English, making it accessible to a broad international community of creators. Whether you're writing detailed instructions in either language, the model can interpret your intent and apply the corresponding edits to your images.

Wan 2.7 Pro Edit offers a range of creative controls that let you fine-tune your results. You can choose from several output size presets including square, square HD, portrait (in both 4:3 and 16:9 ratios), and landscape (in both 4:3 and 16:9 ratios), or specify custom dimensions for your output. This flexibility in aspect ratio and resolution makes the model well-suited for a wide variety of creative contexts — from social media content that demands specific proportions to print-ready artwork that requires particular dimensions. The maximum supported resolution extends up to 14,142 pixels on either side, giving you room for high-resolution output.

A built-in prompt expansion feature is enabled by default, which intelligently enhances your text instructions to help the model better understand and execute your creative vision. This means that even relatively brief or simple prompts can yield rich, detailed results, as the system works behind the scenes to flesh out your intent. If you prefer to maintain exact control over your prompt without any automatic enhancement, you can turn this feature off.

You also have the ability to specify what you don't want in your results through a negative prompt — a description of elements, styles, or artifacts you'd like the model to avoid. This is limited to 500 characters and acts as a guardrail to steer the output away from unwanted directions. For example, you might instruct the model to avoid blurriness, certain color palettes, or specific visual elements that don't fit your creative brief.

For workflows that demand consistency and reproducibility, the model supports a seed value. By using the same seed number across multiple generations with identical settings, you can reproduce the same result — an invaluable feature when iterating on a design, collaborating with a team, or building a series of visually cohesive edits. You can also generate up to four image variations in a single session, allowing you to quickly explore different interpretations of your edit and choose the one that best fits your needs.

Content moderation is built into the model and enabled by default, checking both your input images and the generated output to help ensure responsible use.

Wan 2.7 Pro Edit is ideal for a wide range of creative professionals. Illustrators and fine artists can use it to experiment with style transfers — turning photographs into paintings, sketches, or other artistic interpretations. Graphic designers can quickly prototype visual concepts, test color schemes, or adapt existing assets for new contexts. Photographers can explore dramatic re-edits of their work, transforming mood, setting, or style through simple text descriptions. Content creators producing work for social media, marketing campaigns, or editorial projects will appreciate the speed and flexibility of text-driven editing, which can dramatically accelerate the ideation and production process. Filmmakers and visual storytellers can use the model for rapid concept art iteration and visual development.

The model's ability to work with multiple reference images simultaneously is particularly powerful for composite work and cross-referencing creative elements. Instead of laboriously masking, layering, and blending in traditional software, you can describe your desired combination in natural language and let the model synthesize the result.

In summary, Wan 2.7 Pro Edit brings professional-grade, text-driven image editing to your creative toolkit. With bilingual prompt support, multi-image referencing, flexible output sizing, prompt expansion, negative prompts, reproducible seeds, and batch generation, it provides a comprehensive and accessible set of tools for transforming your images with precision and creative freedom.