Wan 2.5 Image API

Text to ImageImage to Image

AI image generation with editing support using 1-3 reference images.

·Features·FAQ

Documentation

Prompt *

Text description (max 2000 characters)

Aspect Ratio

Reference Images

1–3 images for editing (max 100MB each)

0/3

Click or drag & dropPNG, JPG, WEBP, GIF · Max 100MB

Negative Prompt

What to avoid (max 500 characters)

Prompt Extend

Smart prompt rewriting (default on)

Seed

Seed for reproducibility

Output

Your generated image will appear here

Features

What Wan 2.5 Image API offers

Text-to-image and image-guided editing

Up to three reference images (optional)

Negative prompts within the documented character limit

Prompt extend (smart rewriting), default on in the playground

Seven aspect ratio presets

Optional seed for reproducibility

Use cases

Built for

Primary

Catalog updates: adjust existing shots with one to three references

Localized creatives: iterate with negatives and prompt extend

Storyboards: lock aspect ratio across frames

Lightweight edits: smaller reference sets than Wan 2.6 or 2.7 image

FAQ

About Wan 2.5 Image API

Zero to three images. Omit references for pure text-to-image.

No. It supports text-only generation and runs that use references to align with existing assets.

Wan 2.6 Image allows up to four references and Wan 2.7 Image adds thinking mode and higher resolution tiers. Stay on 2.5 when three references and this control set are enough.