Wan 3.0 Prime image-to-video animates an opening frame into 2 to 30 seconds with generated audio, at 480p, 720p or 1080p. An optional last frame pins the ending, and the prompt only needs to describe motion.
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
masquerade-staircase `image_url` and `last_image_url` are two stills of the same woman on the same marble staircase from the same camera position: masked at the top to start, a few steps lower with the mask in her hand to end. The prompt describes only what happens between them — the mask lifted away, the gown gathered, the slow descent through raised glasses — and the clip opens on the first still and lands on the second with the crowd, the chandelier and the candles unchanged. `duration_seconds` `5` at `720p`, sound on.
- 720p
- audio
- 5s
- Output10s
empty-floor-to-full-house The clip starts empty and has to end full, and `last_image_url` is what makes that safe to ask for: the final frame is fixed in advance, so the model is inventing how a room fills, not what it fills up to. Everything accumulating in between — bodies, light, noise — is left to the prompt. Approve the landing frame as a still first and the long middle becomes far cheaper to iterate on.
- 1080p
- 3:4
- 10s
- Output5s
sauna-ladle-steam The opening frame already holds the set and the wardrobe, so the whole request goes on action and sound: water hitting hot stones, the hiss, cedar creaking, slow breathing. She holds the lens straight through the steam burst, which is the part a still can never show. The frame is yours to host — the URL here is a placeholder.
- 720p
- audio
- 5s
- Output5s
beach-gallop-past The opposite end from the shot above: a fast tracking move alongside a gallop, sand and shallow water thrown up behind the hooves, the low sun flaring straight across the lens. At this speed the opening frame stops helping and the prompt has to carry the physics on its own.
- 720p
- 5s
This example starts at $0.061 per output second. The Playground updates the estimate when you change the settings.
