Private AI
WAN 2.7 image model for text-to-image and multi-image editing. Supports prompt-driven generation, up to 9 reference images, and up to 2K output.
Added Apr 1, 2026
Approx. Price
$0.030 per image
Model Type
both
Generation controls available for this model.
Images Per Run
Up to 4
Output images
Input Images
Up to 9
Reference/edit images accepted • Route max 30 MB
Output Sizes
Negative Prompt
Default
N/A
Details you don't want in the image.
Number of Images
Default
1
Prompt Expansion
Default
No
Let Alibaba expand the prompt before generation.
Resolution
Default
1024*1024
Options (7)
1024*1024 (Square (1:1)), 1280*720 (Landscape (16:9)), 720*1280 (Portrait (9:16)), 1536*1024 (Landscape (3:2)) +3 more
Watermark
Default
No
Add an AI generation watermark to the output image.
Human preference benchmarks sourced from LMArena.
Text to Image
#39 / 64
Arena Score
1098.8
Votes
28,795
Confidence Interval
1093.8 - 1103.8
Image Edit
#15 / 47
Arena Score
1302.2
Votes
40,897
Confidence Interval
1298.3 - 1306.1
Published 2026-05-26 · Matched as wan2.7-image
LMArena DatasetLoading examples…