Alibaba · video
Wan 3.0
Alibaba's Wan 3.0: up to 30 seconds with native audio, frames and up to 10 reference images.
Text to videoImage to videoStart and end frameReferencesNative audio
- Length
- 2 to 30 s
- Resolution
- 480p, 720p, 1080p
- Aspect ratio
- 16:9, 9:16, 1:1, 4:3, 3:4
- References
- Up to 10 image files
About Wan 3.0
Wan 3.0 is the base tier of Alibaba's Wan 3.0, priced at $0.13 per second. It runs as text to video, image to video, start and end frame and references to video, with sound, and its length dial goes from 2 seconds to 30 seconds, further than the earlier Wan models here. Output is 480p, 720p and 1080p, with the shape set automatically or to 16:9, 9:16, 1:1, 4:3 and 3:4.
For the same controls at a higher tier, see Wan 3.0 Prime. A typical generation takes about 4 min.
Good for
- Long Wan shots with sound, up to 30 seconds in one generation
- Keeping a subject consistent from up to 10 reference images
Getting good results
- In reference mode, address the images by their position in the prompt, as the provider shows: Image followed by its number.
- The provider rewrites the prompt before generating, and warns that turning that off tends to lower quality; MegaScenes leaves it on.
- Start at 480p to test a long shot: $0.63 against $2.50 at 1080p.
Wan 3.0 questions
Can I switch the sound off?
Yes. Audio is generated with the video by default and can be turned off; the price is the same either way.
What does the longest clip cost?
$15.00 for 30 seconds at 1080p.