Create a video task (Alibaba Model Studio shape)
The vendor-shaped entry point for Alibaba Model Studio video generation. If you already have DashScope code, point its base_url at Amux; the request and response shapes are the ones your client already speaks.
This endpoint is asynchronous: it returns a task ID and the result is fetched by polling. That is the vendor's own contract rather than a limit imposed here — a 30-second clip takes several to fifteen minutes, longer than any HTTP connection can be held reliably.
The task ID is an Amux ID
The returned task_id is an Amux task ID, not the upstream one. The upstream handle is used internally only: it would reveal which provider served the call, and it is not accepted on any Amux endpoint.
One consequence needs stating explicitly: every retrieval endpoint accepts only this ID. Both GET /v1/tasks/{id} and the Model Studio-shaped alias below read the same task, so an SDK that builds its own polling URL continues to work.
Billing is identical to /v1/tasks
Routing, price book and the reservation held at submit time are all the same. Choosing this endpoint over the unified one changes the wire shape and nothing else.
X-DashScope-Async: enable is accepted and ignored: this endpoint is always asynchronous, so omitting the header is not an error as it is upstream.
Request
- Base URL: https://gateway.amux.ai
- URL: https://gateway.amux.ai/api/v1/services/aigc/video-generation/video-synthesis
- Auth: HTTP bearer
Request body
Example request
{
"model": "alibaba/wan3.0-video",
"input": {
"prompt": "A ginger cat running through fresh snow, slow motion"
},
"parameters": {
"resolution": "1080P",
"duration": 5
}
}Response
Task accepted. Poll GET /v1/tasks/{id} (or the alias below) with output.task_id.
Changes
No recorded changes to this operation across all 1 revision of this API.