Input

prompt *

Text prompts for image generation. Supports both Chinese and English, with a minimum of 1 characters and a maximum of 5,000 characters.

input_urls

Click to upload or drag and drop

Supported formats: JPEG, PNG, WEBP, JPG Maximum file size: 10MB; Maximum files: 9

Upload an image file to use as input for the API

bbox_list

Draw edit regions on each image. Coordinates are saved as [x1, y1, x2, y2] using the original image pixel coordinates. Each image supports up to 2 boxes.

0 boxes selected

Upload image(s) first, then add boxes for each image.

When the group image mode is turned off, n represents the number of generated images (1-4); when it is turned on, n represents the maximum number of images (1-12), and the actual value is determined by the model.

enable_sequential

Enable Gallery Mode

color_palette

Optional custom theme with 3 to 10 colors. Recommended: 8.

0 colors / 0.00%

Add 3 to 10 colors. Ratios must sum to 100.00%.

resolution

Resolution of the generated image. Higher resolutions take longer to generate.(4K generation is available only for text-to-image in Standard Mode)

thinking_mode

Enhances reasoning for higher-quality outputs. Only available when Gallery Mode is off and no images are uploaded. May increase generation time.

aspect_ratio

Aspect ratio of the generated image

watermark

Whether to add a watermark identifier

seed

Using the same seeds and the same prompts for the same version of the model can keep the generated content relatively stable.

Output

output typeimage

Examples

Explore different use cases and parameter configurations

README

Complete guide to using

Affordable Wan2.7-Image API on Kie.ai

Use Kie.ai to access Affordable Wan2.7-Image API for text-to-image, image editing, multi-image control, and high-quality Alibaba visual generation.

Choose the Right Alibaba Wan2.7 Image API for Different Creative Workflows

Alibaba’s Wan 2.7 image family includes two model variants designed for different types of visual workflows: Wan2.7 Image API for faster, more flexible everyday generation and editing, and Wan2.7 Image Pro API for higher-resolution, more quality-focused output. Both support text to image generation, image editor tasks, multi-image input, and thinking mode in supported generation scenarios, making them suitable for teams that need both image creation and image editing in one API flow.

Wan2.7 Image API (wan2.7-image)

Wan2.7 Image API is the standard Alibaba Wan 2.7 image model for teams that need flexible text to image generation and practical image editor capabilities in one API. It supports image generation up to 2K resolution, accepts up to 9 input images for editing or multi-reference fusion, and includes thinking mode in supported text-to-image workflows to improve prompt understanding and output quality. This makes wan2.7-image a strong option for creative testing, product imagery, marketing visuals, and other day-to-day production tasks that require speed and broad editing support.

Wan2.7 Image Pro API (wan2.7-image-pro)

Wan2.7 Image Pro API is the more advanced variant for teams that need higher-quality output from the Alibaba Wan 2.7 image family. Like the standard model, wan2.7-image-pro supports thinking mode, multi-image editing, and flexible prompt-based generation, while extending text-to-image output up to 4K resolution for more demanding visual workflows. That 4K support applies to text-to-image only, while image editing and other image-input scenarios remain below that level. For production design, detailed campaign assets, polished brand visuals, and other quality-sensitive use cases, Wan2.7 Image Pro API is the stronger choice.

Key Features of Alibaba Wan 2.7 Image API

Unified Image Generation and Editing with Wan2.7-Image API

Wan2.7-Image API is built as a unified model for image generation and editing, giving developers a more streamlined way to handle image creation, image transformation, and instruction-based visual control in one workflow. Instead of separating text to image generation from later editing steps, Wan2.7-Image API supports prompt-based generation, reference-based modification, and practical image editor tasks within the same API flow. This unified design makes it more suitable for products that need both original image creation and follow-up refinement, rather than treating generation and editing as disconnected functions.

Portrait Customization and Realistic Face Control in Wan 2.7 Image API

Wan 2.7 Image API offers more granular portrait control for teams that need realistic face generation instead of repetitive, template-like AI portraits. With support for shaping bone structure, eye details, facial contour, makeup, hairstyle, and accessories, Wan 2.7 Image API is better suited to portrait tools, avatar products, beauty-related imaging workflows, and other face-focused creative applications. That added level of portrait customization helps developers build experiences where faces feel more intentional, visually distinct, and better aligned with different identities, ages, and styling directions.

Precise Color Palette Control in Wan2.7 Image Pro API

Wan2.7 Image Pro API is well suited to workflows that need more deliberate visual control instead of approximate styling. It supports precise color handling through reference-based palette extraction and structured HEX-based palette definition, making Wan2.7 Image Pro API more useful for product imagery, campaign assets, illustration systems, and broader brand workflows where color consistency matters. For teams building design-sensitive or commercial visual applications, this gives a more controlled way to guide how color is distributed inside the final output rather than leaving the result to loose prompt interpretation alone.

Advanced Text Rendering in Alibaba Wan2.7 Image API

Alibaba Wan2.7 Image API supports advanced text rendering for more demanding visual tasks that include long-form written content and structured layouts. With support for long text across 12 languages, it is more capable in scenarios such as charts, formulas, reports, infographics, tables, and mixed text-image compositions where readability matters. This makes Alibaba Wan2.7 Image API more useful for developers building content-generation products that need text inside images to remain clearer, more stable, and more usable in practical design workflows.

Multi-Image Editing and Interactive Control with Alibaba Wan2.7 Image API

Alibaba Wan2.7 Image API supports advanced editing workflows that go beyond simple prompt-based image changes. It can work with up to 9 input images for multi-image editing, reference fusion, and more structured creative transformation, making it more suitable for composite generation and guided visual editing tasks. It also supports interactive editing through region-based box selection and natural language instructions, allowing developers to build more precise image editor experiences for targeted object replacement, text changes, layout adjustment, foreground extraction, and other localized edits.

Flexible 2K and 4K Output plus 12-Image Consistency Across Alibaba Wan2.7 Image API

Alibaba Wan2.7 Image API provides two model variants for different production priorities while also supporting more consistent multi-image output. Wan2.7 Image API supports generation up to 2K, while Wan2.7 Image Pro API supports up to 4K for text-to-image in supported scenarios, giving developers a clearer way to match model choice to output requirements. At the same time, Alibaba Wan2.7 Image API can generate up to 12 stylistically consistent images in one request, which makes it especially useful for sequential storytelling, storyboards, product catalogs, children’s books, wedding series, slide visuals, and other workflows that need stronger subject and style consistency across multiple outputs.

Core Input Options and Model Capabilities Across Wan2.7 Image API and Wan2.7 Image Pro API

Both Wan2.7 Image API and Wan2.7 Image Pro API share a closely aligned input structure for generation and editing workflows, which makes it easier to move between the standard and Pro variants without changing the overall integration logic too much. The main differences are concentrated in output size and generation range, while both models continue to support multi-image input, thinking mode in eligible scenarios, and interactive editing controls.

Parameter	Wan2.7 Image API	Wan2.7 Image Pro API
Input images	Up to 9 input images	Up to 9 input images
Output size	1K, 2K, and supported custom dimensions	1K, 2K, 4K for text-to-image, and supported custom dimensions
Number of outputs	1–4 outputs, or up to 12 when image set mode is enabled	1–4 outputs, or up to 12 when image set mode is enabled
Thinking mode	in eligible text-to-image scenarios	in eligible text-to-image scenarios
Interactive editing boxes	✔	✔

How to Get Started with Wan2.7-Image API on Kie.ai

Get started with our product in just a few simple steps...

Sign Up or Log In to Kie.ai and Get Your Wan2.7-Image API Key

Create your account or sign in to Kie.ai first, then generate your Wan2.7-Image API Key from the API access area. This key is required before calling either wan2.7-image or wan2.7-image-pro, and it is the starting point for both playground testing and later deployment. Once the key is ready, you can begin working with the same request structure across generation and editing workflows.

Test Wan2.7-Image API Free in the Playground

Use the playground to test Wan2.7-Image API before deployment, including both the standard and Pro variants. This is the best place to compare text-to-image generation, image-to-image editing, multi-image input behavior, thinking mode, and different output settings without committing to a full integration first. A playground workflow also makes it easier to evaluate when wan2.7-image is sufficient for faster everyday generation and when wan2.7-image-pro is the better fit for higher-resolution text-to-image output or more quality-sensitive visual tasks.

Configure Your Wan2.7-Image API Request for the Right Workflow

After testing, set up the request parameters based on the workflow you want to deploy. For pure generation, define the prompt, choose the model variant, and set the output size that matches your delivery needs. For editing workflows, attach one or more input images and add the editing instruction in the prompt. If you need more structured visual output, you can also configure image count, interactive editing regions, or image set generation depending on the scenario.

Integrate Wan2.7-Image API into Your Application or Workflow

Once the request format is confirmed, connect Wan2.7-Image API to your application, automation flow, creative tool, or internal content pipeline. At this stage, the goal is usually to turn successful playground tests into repeatable API calls for production use, whether that means generating marketing visuals, powering an image editor, building portrait tools, or supporting multi-image creative workflows. Because both model variants share a closely aligned input structure, it is easier to move between standard and Pro usage as your output needs change.

Deploy Wan2.7-Image API in Production and Optimize as Needed

After integration is stable, deploy Wan2.7-Image API into production and refine the model setup based on actual usage. Teams typically optimize around output size, generation mode, image inputs, and model selection, using wan2.7-image for faster general-purpose workflows and wan2.7-image-pro for more demanding text-to-image quality requirements. From there, the deployment can be expanded to support broader generation, editing, and consistent multi-image workflows as product needs grow.

Compare Wan2.7-Image Across Text-to-Image and Image Editing

In anonymized human preference tests, Wan2.7-Image outperformed several leading industry models by delivering stronger visual fidelity, more accurate text rendering, and a better understanding of complex visual concepts across both generation and editing tasks.

More Competitive Text-to-Image Generation with Wan2.7-Image

In the text-to-image generation comparison, Wan2.7-Image shows a notably balanced profile across photorealistic imaging, art & design, text rendering, layout & info visualization, world knowledge, and complex semantics & reasoning. Compared with GPT Image 1.5 and Seedream 4.5, it appears stronger in text rendering and more capable in structured or knowledge-aware image tasks, while still remaining competitive in photorealistic and design-oriented output. Compared with Kling Image 3.0 Omni and Seedream 5.0 Lite, the advantage looks even clearer across text-heavy, layout-sensitive, and reasoning-related dimensions. Nano Banana Pro still leads the chart overall, but Wan2.7-Image stands out as a stronger all-around model than several other competitors rather than leaning too heavily on only one generation strength.

More Well-Rounded General Image Editing in Wan2.7-Image

In general image editing, Wan2.7-Image again shows a more even performance profile across the main editing dimensions in the chart, including identity & reference, advanced editing, multi-image editing, local editing, text editing, object & scene editing, and style & lighting editing. While Nano Banana Pro remains ahead overall, Wan2.7-Image compares favorably with Seedream 5.0 Lite across nearly every editing category shown. Relative to Seedream 5.0 Lite, it looks stronger in multi-image editing, local editing, text editing, and object & scene editing, while staying close in advanced editing and style & lighting editing. That makes Wan2.7-Image a more practical choice for teams that need one model to handle reference-guided edits, localized revisions, text updates, and broader multi-image transformation workflows without being limited to a narrower editing specialty.

How Alibaba Wan2.7 Image API Fits Real Creative Workflows

Faster Campaign Visuals and Creative Testing with Wan2.7-Image API

Wan2.7-Image API works well in fast-moving marketing and creative workflows where teams need to turn ideas into visuals without a long design cycle. It can be used for ad concepts, landing page graphics, social media images, promotional banners, and other campaign assets that start from a prompt and often need quick follow-up edits. Because Wan2.7-Image API combines generation and editing in one flow, it is especially practical for teams that want to move from rough visual exploration to more usable creative output without switching across multiple image tools.

More Controlled Portrait and Avatar Creation in Wan 2.7 Image API

Wan 2.7 Image API is a strong choice for portrait-focused products that need faces to feel more intentional and less generic. With more detailed control over bone structure, eye shape, contour, makeup, hairstyle, and accessories, Wan 2.7 Image API is well suited to avatar tools, profile image generation, beauty-related imaging experiences, and other portrait workflows where facial customization matters. This gives developers a better foundation for face-centered products that depend on more realistic, more varied, and more identity-sensitive visual output.

Higher-End Brand Assets and Design Delivery with Wan2.7 Image Pro API

Wan2.7 Image Pro API is better aligned with visual workflows where output quality carries more weight than raw iteration speed. It is a stronger fit for premium campaign creatives, product marketing assets, branded illustrations, presentation visuals, and other design-heavy content that benefits from more polished text-to-image generation. Since Wan2.7 Image Pro API supports up to 4K in supported text-to-image scenarios, it is especially useful when the final visual needs to feel more refined for production-facing delivery rather than quick internal concepting.

Storyboards, Catalogs, and Structured Visual Sequences with Wan2.7 Image API

Wan2.7 Image API is especially useful for workflows that depend on consistency across multiple outputs rather than one-off image generation. It can support storyboards, product catalogs, children’s books, wedding series, slide visuals, and other sequential content where the subject, styling, or scene logic needs to stay more coherent from image to image. With support for up to 12 consistent images in one generation and up to 9 reference images for more guided outputs, Wan2.7 Image API is a strong match for teams building visual storytelling tools and multi-image production workflows.

Why Kie.ai Is a Better Choice for Wan 2.7 Image API Access

Affordable Wan2.7-Image API Pricing with Practical Model Access

Kie.ai offers a more practical way to adopt Wan2.7-Image API for teams that want strong image generation and editing capabilities without turning API adoption into an unnecessarily heavy commitment. The advantage is not only in Affordable Wan2.7-Image API Pricing, but in pairing that value with usable access to both standard and Pro workflows, so teams can evaluate the model more efficiently for ongoing creative production, product integration, and commercial use.

Complete Wan2.7-Image API Documentation for Faster Integration

Kie.ai is also a stronger choice because complete Wan2.7-Image API Documentation makes the model much easier to move from testing into real implementation. Clear documentation helps developers understand the core request structure, supported generation and editing workflows, model differences, and practical integration logic more quickly. For teams comparing providers, strong Wan2.7-Image API Documentation can make the difference between a model that looks capable in theory and one that is actually efficient to deploy in practice.

Dependable 24/7 Service for Ongoing Wan2.7-Image API Use

For production-facing image workflows, reliable support matters as much as generation quality. 24/7 service gives teams more confidence when using Wan2.7-Image API for live products, internal tools, and commercial creative pipelines that cannot afford unnecessary delays. Whether the need is ongoing integration help, deployment troubleshooting, or faster issue response during active usage, round-the-clock service makes Wan2.7-Image API a more dependable choice for developers and businesses that need stable long-term API support.