alibaba/wan-2.6/reference-to-videoWan 2.6 reference-to-video from Alibaba casts people, animals or objects from one to three reference clips into a new scene described by your prompt. Clips run 2 to 10 seconds at 720p or 1080p, in a single take or multi-shot.
Filtering differs by endpoint: Image to video — Uncensored, Text to video — Uncensored, Reference to video — Filtered.
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
bottle-new-set One product clip as the reference, re-staged in a new room while the camera repeats its slow circle.
- 720p
- 16:9
- 5s
- Output5s
cast-three-clips The references here are clips rather than stills: `reference_video_urls` takes up to three, and the brief casts two people and a street out of them into a scene that never existed. Motion references carry what a photograph cannot — how someone moves, how a place behaves — which is why clips are the right input when the new scene has to feel continuous with the old footage. The result runs 2 to 10 seconds.
- 720p
- 5s
This example starts at $0.065 per output second. The Playground updates the estimate when you change the settings.
