What Wan 3.0 Omni Reference is good at
Video models are billed per second or per clip: turn a description or a picture into a clip a few seconds long.
Mix images, video, audio and text as references to reuse characters, products, scenes and voices. Outputs 1080P video with sound up to 30 s (including reference video).
Good for
- Short video ads and product showcases
- Bringing a still image to life
- Short clips for social media
Less suited for
- Videos several minutes long: generate in parts and edit together
- Articles or code: use a chat model
Code sample
How to call it will be in the docs. You can sign up and create an API key now; you will not need a new key when it opens.
What it costs, roughly
Estimated from the unit price above; you are charged per generation.
| One 5-second clip5 × Rp 1.792 | Rp 8.960 |
| One 10-second clip10 × Rp 1.792 | Rp 17.920 |
| A 30-second ad (generated in parts)30 × Rp 1.792 | Rp 53.760 |
| A minimum Rp 50.000 top-up coversbilled per second | ≈ 27 s |
Model status
Calls over the last 24 hours.
Related models
One API key calls them all; to switch, just change the model name.
Other Alibaba models
Wan 3.0Coming soonRp 932 per second
HappyHorse ReferenceComing soonRp 3.117 per second
Wan 3.0 Text-to-VideoComing soonRp 1.792 per second
Wan 3.0 ReferenceComing soonRp 1.792 per second
HappyHorse Video EditComing soonRp 3.117 per second
HappyHorse 1.1 ReferenceComing soonRp 2.986 per second
Alternatives at a similar price
Kling AvatarComing soonRp 1.792 per second
Hailuo H3 Max First-Last FrameComing soonRp 1.971 per second
Hailuo H3 Max Text-to-VideoComing soonRp 1.971 per second
MiniMax H3 MaxComing soonRp 1.971 per second