Wan 2.7 Image
What is Wan 2.7 Image?
Wan 2.7 Image is the faster of the two Wan 2.7 image variants. It supports text-to-image, editing, and image-guided generation with up to nine reference images, seven aspect ratios, and a thinking mode that adds reasoning on text-only generations. Output goes up to 2K. Prompts cap at 5000 characters and editing preserves the input aspect ratio. Use it when you need the Wan 2.7 image upgrades but want shorter turnaround than the Pro variant.
Key features of Wan 2.7 Image
Four features cover what changes versus Wan 2.6 Image.
Up to 9 reference images
More than double Wan 2.6 Image's four-reference cap. Useful for brand toolkits, multi-character casts, or complex composite briefs that need many style anchors in one call.
Thinking mode for text-only runs
`thinking_mode: true` adds a reasoning step before drawing. On by default. Useful for dense scenes, multi-subject layouts, and prompts where composition matters as much as the look.
Up to 2K output
Wan 2.7 Image targets 2K output for web and social. For 4K masters, switch to Wan 2.7 Image Pro on the same parameter shape.
5000-character prompts
Two and a half times the Wan 2.6 Image cap. Useful for detailed brand briefs and multi-element scene prompts that need every constraint stated.
Best for
Production pipelines
Faster turnaround while staying on Wan 2.7 Image. Good for high-volume creative ops.
Web and social
2K assets for screens, responsive layouts, and feed posts.
Multi-reference creative
Combine up to nine images for style or subject alignment in a single call.
Difficult prompts
Thinking mode handles dense scenes and multi-subject layouts where Wan 2.6 Image plateaus.
Brand toolkits
Many references in a single request keeps colour, character, and layout consistent across a series.
Iterative editing
Reproducible runs with seed and prompt extend control across rounds of feedback.
API examples
Call Wan 2.7 Image from any language by POSTing to /v1/tasks. Full parameter docs live at docs.unifically.com/models/image/alibaba/wan-2.7-image.
curl -X POST https://api.unifically.com/v1/tasks \
-H "Content-Type: application/json" \
-H "Authorization: Bearer YOUR_API_KEY" \
-d '{
"model": "alibaba/wan-2.7-image",
"input": {
"prompt": "A cozy cabin in the mountains surrounded by pine trees",
"aspect_ratio": "16:9"
}
}'
Successful submission returns a task_id. Poll GET /v1/tasks/<task_id> or set a callback_url on the request to receive the finished result.
FAQs
People also ask
Wan 2.7 Image is the faster of the two Wan 2.7 image variants on Unifically. It supports text-to-image, editing, and image-guided generation with up to nine reference images, seven aspect ratios, and a thinking mode for harder text-only prompts. Output goes up to 2K. Prompts are capped at 5000 characters.
Both variants share the same parameter shape (up to nine references, thinking mode, seven aspect ratios). Wan 2.7 Image runs faster and outputs up to 2K. Wan 2.7 Image Pro outputs up to 4K and is the higher-quality variant.
Yes. Pass image_urls with up to nine images. When references are present, the request follows the editing path. Editing preserves the input aspect ratio.
Thinking mode (thinking_mode: true) spends extra reasoning before drawing on text-only generations. It is on by default and is not used when reference images are supplied.
Up to nine. Useful for brand toolkits, complex character references, and multi-style alignment.
Move to Pro when you need 4K output for print, packaging, OOH, or large-format work. Stay on the standard Wan 2.7 Image when 2K is enough and turnaround matters.
Yes. Pass an integer seed alongside the prompt and any references. Re-sending the same prompt, references, and seed returns the same image.
