Alibaba
Wan 2.7 reference to video
Build a new scene around up to five image or video references. Name the references in the prompt to direct the subjects, setting or voice that the model should carry over.
API pricing · From $0.10/output secAPI model ID
alibaba/wan-2-7-r2vSource inputs
Best for
- Putting a recurring character in a new scene
- Directing a shot with separate subject and location references
Technical details
Read the model documentation- API model ID
alibaba/wan-2-7-r2v- Provider
- Alibaba
- Modality
- Video
- Input
- Text prompt · Images · Video
- Output
- Video
- Endpoint
POST /v1/alibaba/generations- Resolution
- 720p or 1080p
- Duration
- 2–15 seconds, up to 10 seconds with a reference video
- References
- Up to 5 images and videos combined
- API constraints
- A separate voice per reference is not exposed.
Request parameters
promptresolutiondurationaspect_ratiofirst_frame_url
Additional parameters
negative_promptprompt_extendreference_image_urlsreference_video_urlsseedwatermark
Pricing
| Billable unit and options | Rate (USD) |
|---|---|
| Per second of output video720preference video seconds are billed too | $0.10 |
| Per second of output video1080preference video seconds are billed too | $0.15 |
Billing notes
- Billed on the seconds and resolution the generator reports for the finished clip.
- Leaving out resolution gives 1080p.
Catalogue rates on 21 Nov 2026. Prices exclude VAT.
View in pricing catalogue