Brand and marketing teams
Generate visual campaigns with stronger text rendering, specified brand colors, and multiple related assets from one creative direction.
Independent tool overview
Wan2.7-Image at a glance
Wan2.7-Image is Alibaba's unified image model family for text-to-image generation, instruction-based editing, multi-image composition, precise region edits, and coherent image sets.
Visit the official Wan2.7-Image site ↗
- Best for
- Campaign images, product visuals, character-consistent sets, and iterative editing
- Models
- wan2.7-image and wan2.7-image-pro
- Maximum resolution
- 2K standard; 4K Pro for text-to-image only
- Reference images
- Up to 9
- Image-set output
- Up to 12 related images
- International API price
- $0.03 standard or $0.075 Pro per successful image
Overview
What Wan2.7-Image is
Wan2.7-Image is Alibaba's current image generation and editing family. The same API can create an image from text, edit an existing image, combine multiple references, and generate a sequence of related images, reducing the need to switch between separate generation and editing models.
There are two versions. wan2.7-image is the faster, lower-cost option with output up to 2K. wan2.7-image-pro costs more and extends text-to-image generation to 4K, while editing and image-set work still tops out at 2K. Both models emphasize text rendering, subject consistency, brand-color control, and complex instruction following.
Wan2.7-Image is available through Alibaba Cloud Model Studio and Wan's product experience. API users must choose a supported region and use the matching endpoint and API key. The pricing below is Model Studio API pricing; any consumer credits or promotions shown inside Wan can differ.
Use cases
Who Wan2.7-Image is best for
The strongest fit depends on the job you need the product to complete, not the size of its feature list.
Character-consistent storytelling
Use multiple references and sequential generation to create related scenes while preserving a subject across an image set.
Developers building image workflows
Access generation, editing, regional edits, and image sets through synchronous or asynchronous Alibaba Cloud Model Studio APIs.
Capabilities
Core Wan2.7-Image features
Unified generation and editing
Use one model family for text-to-image, single-image editing, multi-image composition, style changes, and reference-guided generation.
Standard and Pro models
Choose the faster, less expensive standard model for up to 2K output or Pro when 4K text-to-image generation is worth the higher price.
Multi-image references
Provide as many as nine input images and refer to them by order when combining subjects, objects, environments, or visual styles.
Precise region editing
Target selected areas with bounding boxes—up to two boxes per input image—instead of asking the model to infer the edit location from text alone.
Sequential image sets
Generate up to 12 related images from text or image references for storyboards, campaigns, product sequences, and other coherent series.
Character and subject consistency
The model is designed to preserve referenced subjects across edits and multi-image sets more reliably than a prompt-only workflow.
Brand-color control
Pro supports palette guidance for workflows that need closer alignment with a defined set of colors.
Flexible dimensions
Use square resolution shortcuts or custom dimensions across aspect ratios from 1:8 to 8:1, within each model and task's pixel limits.
Process
How the Wan2.7-Image workflow works
Step 1
Choose standard or Pro
Use wan2.7-image for faster 2K work or wan2.7-image-pro for higher-quality workflows and 4K text-to-image output.
Step 2
Select a task
Start from text, supply one or more images for editing or composition, or enable sequential generation for a related image set.
Step 3
Write the instruction
Describe the subject, composition, text, colors, and intended changes; identify reference images by their input order when needed.
Step 4
Set output controls
Choose resolution, dimensions, number of images, thinking mode, watermark behavior, and bounding boxes for precise edits.
Step 5
Save and review
Download successful API outputs promptly because Alibaba's temporary image URLs expire after 24 hours, then review text, identity, and brand details.
Cost
Wan2.7-Image pricing and free plan
Alibaba Cloud Model Studio bills Wan2.7-Image per successfully generated image; input images and failed generations are not billed. International access in Singapore lists 50 free images per model for eligible new users, valid for 90 days. Prices inside Wan's consumer experience or limited-time Model Studio promotions may differ.
wan2.7-image
$0.03 per image
International Model Studio API price for the faster standard model.
- Up to 2K generation and editing
- 50-image new-user quota in the eligible international region
- Only successful output images are billed
wan2.7-image-pro
$0.075 per image
International Model Studio API price for Pro.
- Up to 4K for text-to-image
- Editing and sequential image sets remain limited to 2K
- 50-image new-user quota in the eligible international region
China mainland API
$0.028671 standard / $0.068761 Pro
Published China (Beijing) prices per successful image.
- Requires a China-region Model Studio setup
- Regional API keys and endpoints are not interchangeable
Fine-tuning
$0.015 per 1,000 training tokens
Published SFT-LoRA training price for either Wan2.7 image model in the supported Singapore workflow.
- Custom deployment itself is listed as free
- Invocations use the standard base-model per-image price
- Training and inference costs are separate
Pricing checked . Check current pricing at the source ↗
Assessment
Wan2.7-Image strengths and limitations
Where it stands out
- One model family covers generation, editing, multi-reference composition, and related image sets
- Pro offers true 4K output for text-to-image generation
- Up to nine reference images support more complex composition and consistency workflows
- Bounding-box controls make localized edits more precise
- Sequential mode can produce as many as 12 coherent images in one request
- Clear pay-per-successful-image API pricing avoids charging for failed outputs
What to consider
- Pro's 4K capability applies only to text-to-image; editing and image sets are capped at 2K
- The standard model is limited to 2K output
- Regional models, endpoints, and API keys must match and are not interchangeable
- Within each model, supported resolutions use the same per-image rate; Pro's 4K text-to-image output is billed at the standard Pro rate
- Sequential mode disables some controls, including thinking mode and color palette
- Generated image URLs expire after 24 hours and need to be copied to permanent storage
- Prompts and outputs still require human review for text accuracy, likeness rights, trademarks, copyright, and brand compliance
Compare
Wan2.7-Image alternatives
The right alternative depends on the specific output, workflow, controls and budget your project requires.
Design
Qwen Image 2.0
Choose Qwen Image 2.0 when negative prompting or generating up to six independent variants per call matters more than Wan's 4K Pro and 12-image sequential mode.
Explore Qwen Image 2.0 →Content Creator
FLUX.2
Choose FLUX.2 when you want Black Forest Labs' model family, broader deployment choices, or an open-weight Klein option.
Explore FLUX.2 →Design
Ideogram
Choose Ideogram when a polished creator interface and typography-led marketing graphics are more important than Alibaba Cloud API controls.
Explore Ideogram →Content Creator
Nano Banana 2
Choose Nano Banana 2 when Google's Gemini ecosystem, conversational editing, and fast high-volume generation better fit your workflow.
Explore Nano Banana 2 →Questions
Wan2.7-Image FAQs
What is Wan2.7-Image?
Wan2.7-Image is Alibaba's unified image generation and editing family. It supports text-to-image, image editing, multi-image reference generation, precise region edits, and related image sets.
What is the difference between Wan2.7-Image and Wan2.7-Image Pro?
The standard model is faster, costs $0.03 per successful image internationally, and outputs up to 2K. Pro costs $0.075 per image and supports up to 4K for text-to-image; its editing and image-set outputs remain capped at 2K.
How much does Wan2.7-Image cost?
As of August 29, 2026, Alibaba Cloud Model Studio lists international API pricing at $0.03 per successful image for wan2.7-image and $0.075 for wan2.7-image-pro. Consumer credits and promotions may use different prices.
Is there a free Wan2.7-Image quota?
Alibaba Cloud lists 50 free images for each model for eligible new users in its international Singapore service. The published quota is valid for 90 days; check your console because eligibility and promotional terms can change.
Can Wan2.7-Image generate 4K images?
wan2.7-image-pro can generate up to 4K from text. The standard model is limited to 2K, and Pro editing or sequential image-set tasks are also limited to 2K.
How many reference images can Wan2.7-Image use?
Both Wan2.7 image models accept up to nine input images. Prompts can identify references by their position, such as image 1 and image 2.
How many images can Wan2.7-Image generate at once?
A standard generation request can ask for up to four outputs. Sequential image-set mode can request up to 12 related images, although the model may return fewer than the requested maximum.
Does Wan2.7-Image support precise local edits?
Yes. Its interactive editing control accepts bounding boxes for targeted areas, with up to two boxes per input image.
Is Wan2.7-Image the same as Wan 2.7 video?
No. Wan2.7-Image produces still images. Alibaba also offers separate Wan 2.7 video models for text-to-video, image-to-video, reference-to-video, and other motion workflows.
Bottom line
Our Wan2.7-Image verdict
Wan2.7-Image is a strong fit when one production workflow needs both image creation and substantial editing, especially with multiple references or coherent asset sets. The standard model is the value choice for 2K work; Pro makes sense when 4K text-to-image, brand-color control, or maximum quality justifies paying 2.5 times as much. Test both on the same real campaign before standardizing, and budget for permanent output storage and human brand review.
Visit Wan2.7-Image website ↗Similar Design Tools
View all Design AI tools →
Uni-1
Uni-1- Luma's unified model that reasons, generates, and understands across text and images
Claude Design
Claude Design - Anthropic's new design tool for collaborating with Claude to create polished visual work like designs, prototypes, slides, one-pagers, and more.
QuiverAI Arrow 1.0
Arrow 1.0 - QuiverAI's new top-ranked SVG generation model, now in public beta
ChatGPT Images 2.0
ChatGPT Images 2.0 - OpenAI's new SOTA image generation model with next-generation text rendering, realism, and thinking capabilities
AI training for the future of work.
Get access to all our AI courses, hundreds of real-world AI use cases, live expert-led workshops, an exclusive network of AI early adopters, and more.
AI Courses
Get unlimited access to all of our current & upcoming industry-specific AI courses for the duration of your subscription.
Daily Guides
To keep up with the rapid pace of AI, our team publishes AI implementation guides daily. Our library contains 300+ practical use cases to automate real-world work.
Workshops
Join weekly, live, interactive sessions with industry leaders who are at the forefront of AI for hands-on implementation guidance and exclusive insights.
Community
Network with an exclusive community of AI-first professionals who are working smarter with AI. Learn how early adopters are using AI in their work and businesses.