Wan 2.5 text-to-video from Alibaba turns a written shot into a 5 or 10 second clip with generated sound, at 480p, 720p or 1080p. Pick 16:9, 9:16, 1:1, 4:3 or 3:4, and optionally supply an audio track to follow.
Workbench
These examples were written for this endpoint. Replace the prompt and sample media URLs with your own accessible inputs, then check the current quote and model availability before submitting.
- Output5s
bar-slow-dance Two people in one move, from words alone: after closing time she pulls him in by his tie until their faces almost meet, then leans back in his arms and laughs while the camera orbits them. The haze in the amber light beams gives the orbit something to reveal.
- 720p
- 16:9
- 5s
- Output10s
sax-solo-ten-seconds `duration_seconds` is `10`, the longer of the two lengths this endpoint accepts, at `1080p` in `16:9`. The brief asks something specific of the audio pass: the horn's sound matching the fingering, which is a synchronisation claim rather than an ambience one. If you are going to test generated sound, test it against a visible mechanism — then you can see for yourself whether it lines up.
- 1080p
- 16:9
- 10s
- Output5s
harbour-dawn-swim One action and one camera position, with the sound written out so the splash and laugh land on the right frames.
- 720p
- 16:9
- 5s
- Output10s
record-shop-vertical A ten-second vertical take at 1080p: a tilt from hands to face, with a negative prompt keeping captions and extra people out.
- 250924
- 1080p
- 9:16
This example starts at $0.045 per output second. The Playground updates the estimate when you change the settings.