What Wan 3.0 does
Wan 3.0 by Alibaba generates video from text, an image, first and last frames, or a set of references — with native audio. Clips run from 2 to 30 seconds at 480p, 720p, or 1080p, and you can guide a scene with up to 10 reference images and 5 reference videos. Wan 3.0 Prime is the high-speed tier with the same inputs. Wan 2.6 and Wan 2.5 are also available.
- Native Audio
- Wan 3.0 and Wan 3.0 Prime generate video with native audio, so your clip comes out with sound.
- Wan 3.0 Prime: High-Speed Tier
- Wan 3.0 Prime is the high-speed Wan 3.0 tier — the same text, image, frame, and reference inputs when you need results faster.
- Long Takes up to 30 Seconds
- Choose any duration from 2 to 30 seconds at 480p, 720p, or 1080p — room for a full scene in a single generation.
- Multi-Reference Consistency
- Add up to 10 reference images and 5 reference videos to steer the scene and keep characters consistent.
How to use it
Choose Model & Mode
Pick Wan 3.0 or the faster Wan 3.0 Prime, then start from text, an image, first and last frames, or references.
Write Prompt & Upload
Describe your scene with camera direction. If your mode uses references, upload the images or videos.
Generate
Choose a duration from 2 to 30 seconds and a resolution up to 1080p, then generate your video with sound.
Questions
Wan 3.0 is Alibaba's AI video model for text, image, first/last-frame, and multi-reference video with native audio. It generates clips from 2 to 30 seconds at up to 1080p.
Yes. Wan 3.0 and Wan 3.0 Prime generate video with native audio. Wan 2.5 also supports audio, while Wan 2.6 generates video without sound.
Wan 3.0 generates 2 to 30 second videos at 480p, 720p, or 1080p in 16:9, 9:16, 1:1, 4:3, and 3:4. You can add up to 10 reference images and 5 reference videos.
Use Wan 3.0 for long takes with audio and references, and Wan 3.0 Prime when you need the same inputs faster. Wan 2.6 (5–15s, 720p or 1080p, text, image, or video to video) and Wan 2.5 (5 or 10s, 720p or 1080p, with audio) remain available.