Wan 3.0 AI
Alibaba's Omni-Reference Video Model

Wan 3.0 by Alibaba unifies text, image, video, audio, keyframes and document references in a single model. Generate 480p, 720p or 1080p videos up to 30 seconds with native audio — text-to-video and image-to-video with first/last frame control, live on VO3 AI.

Try Wan 3.0 Now View Pricing

↓ Explore Wan 3.0 Features

Generate Video with Wan 3.0

Try Wan 3.0 text-to-video and image-to-video directly. Select Wan 3.0 from the model dropdown to experience Alibaba's latest omni-reference video technology.

🎬

Generate AI Model videos with AI

Resolution:

·Duration:

·Ratio:

0/2000 characters

Try these ideas:

🎬 One video · $2.99 · no sign-up — we’ll email you the link. Already have credits? Sign in

What is Wan 3.0?

Wan 3.0 is Alibaba's most advanced AI video generation model — a major leap from Wan 2.7. Instead of separate endpoints for each task, Wan 3.0 handles text-to-video, image-to-video, and first/last-frame animation through a single unified model, and layers in omni-reference understanding: images, videos, audio, keyframes, documents, and webpages can all shape the scene, motion, sound, visual identity, and narrative direction of the output.

Wan 3.0 is available on VO3 AI through the KIE API infrastructure. It outputs 480p, 720p, or 1080p video with durations up to 30 seconds and native audio built in. On the create page you can use Wan 3.0 today for text-to-video and image-to-video with first and last frame control.

What Wan 3.0 Can Do

Text to Video

Generate 480p/720p/1080p videos with native audio from
a text prompt — English or Chinese

Image to Video

Animate a single image with
first-frame control and adaptive aspect ratio

First → Last Frame

Upload a start and end frame —
Wan 3.0 generates smooth motion between them

Reference Images

Up to 10 reference images
to lock character and style consistency

Reference Video

Up to 5 reference videos to carry
motion and look into a new scene

Reference Audio

Up to 5 audio references to drive rhythm, voice and timing

Document & Webpage

Reference a document or webpage
to shape the narrative direction

Multilingual Prompts

Prompts up to 20,000 characters in English and Chinese

Wan 3.0 Technical Specifications

480p, 720p & 1080p

Three resolution tiers for speed-first
drafts up to cinematic 1080p output

Up to 30s Duration

Longer-form generation with duration-based
pricing for cost control

Native Audio Generation

Built-in audio synthesis that
matches generated visuals automatically

Adaptive Aspect Ratio

Adaptive output plus 16:9, 9:16, 1:1, 4:3 and 3:4 framing

Single Unified Model

One endpoint for text, image,
keyframe and reference generation

Character Consistency

Omni-reference keeps character identity
stable across the scene

Wan 3.0 Pricing

Resolution	5s	10s
720p	4 credits	6 credits
1080p	6 credits	12 credits

Image-to-video uses the same rates · Credits start from $2.99 · View all plans

Frequently Asked Questions About Wan 3.0

Wan 3.0 is Alibaba's latest all-in-one video generation model. Compared to Wan 2.7, it unifies every generation mode behind a single endpoint and adds omni-reference understanding — you can guide a video with text, images, first/last keyframes, reference videos, reference audio, documents, and even webpages. It supports 480p/720p/1080p output, durations up to 30 seconds, and native audio.

Wan 3.0 accepts a text prompt (up to 20,000 characters, English or Chinese), a first frame and optional last frame for keyframe animation, up to 10 reference images, up to 5 reference videos, up to 5 reference audio clips, a reference document, and a reference webpage. On VO3 AI you can use it right now for text-to-video and image-to-video (first/last frame).

Wan 3.0 pricing is based on resolution and duration: 720p 5s costs 4 credits and 10s costs 6 credits; 1080p 5s costs 6 credits and 10s costs 12 credits. Image-to-video uses the same rates.

Yes. Wan 3.0 generates native audio that matches the scene automatically. You can also supply reference audio to guide the sound and timing of the output.

Wan 3.0 supports 480p, 720p, and 1080p output. Duration can range up to 30 seconds; on VO3 AI the create page offers 5s and 10s presets for text-to-video and image-to-video.

Yes. All videos generated on VO3 AI, including Wan 3.0 outputs, can be used for commercial purposes including marketing, social media, advertising, and business use.

Start Creating with Wan 3.0

Omni-reference generation, 480p/720p/1080p output, native audio. Alibaba's most capable AI video model, available on VO3 AI.

Start Creating View Pricing

Wan 2.7 Seedance 2.5 MiniMax H3

Wan 3.0 AI Alibaba's Omni-Reference Video Model