Wan 3.0 — Multimodal AI Video Generator

Direct longer, reference-rich videos with native audio and flexible framing controls.

Video Creation
Configure a Wan 3.0 workflow for text, image, or video-driven generation.

Compose a prompt or guide the result with optional audio, link, or document references.

0 / 20000 characters
seconds

Choose any whole number from 2 to 30 seconds. Smart duration is charged as 30 seconds.

Generate Audio

Include a native audio track in the generated video.

Advanced References & Controls+

Add audio, a public webpage or one document, and set a reproducible seed.

Up to 5 · 1–15s each · 15MB
Select or drag audio references
MP3 or WAV

One public URL with no login required.

One file · up to 100MB
Select or drag a document
PDF, Office, iWork, TXT, or Markdown

Use 0 for a random result, or reuse a seed to improve reproducibility.

Safety checker
Enabled for every generation.
Required: 160 creditsAvailable: 0 credits
Enter a prompt or add a reference before generating.
Insufficient creditsGet more credits
Preview Surface
Review the current mode, source media, and output settings before generating.
Text to Video
Wan 3.0
Your Wan 3.0 prompt stage

Long-form prompting, multimodal references, native audio, and flexible framing are ready for final rendering.

Generated output will appear here once the job completes.Waiting for input