Tencent Hunyuan Video

Endpoints

POST /v1/wand/hunyuan-video/generation
POST /v1/videos/generations
Authorization: Bearer $API_KEY
Content-Type: application/json

Both paths are equivalent and accept the same body. The upstream accepts the task immediately; poll via GET /v1/tasks/{taskId}.

Request body

Field	Type	Required	Description
`model`	string	Yes	Complete public model ID, currently `tencentcloud/hy-video-v1.5`.
`prompt`	string	For text-to-video	Text prompt.
`negative_prompt`	string	No	Negative prompt describing what should not appear.
`image`	string	For image-to-video	Base64-encoded input image (mutually exclusive with `image_url`; takes precedence when both are provided).
`image_url`	string	For image-to-video	Public URL of the input image (mutually exclusive with `image`).
`aspect_ratio`	string	No	Aspect ratio, text-to-video only; ignored for image-to-video (determined by the input image). One of `16:9` (default), `9:16`, `1:1`, `4:3`, `3:4`.
`n`	integer	No	Number of output videos; defaults to `1`.
`duration`	number	No	Video duration in seconds; currently fixed at 5 seconds.
`seed`	integer	No	Random seed (passed through; whether it takes effect depends on the model).
`callback_url`	string	No	Public terminal callback URL handled by the platform.

Providing image or image_url selects image-to-video mode; omitting both selects text-to-video.

Examples

curl "https://api.modelrover.ai/v1/wand/hunyuan-video/generation" \
-H "Authorization: Bearer $API_KEY" \
-H "Content-Type: application/json" \
-d '{
  "model": "tencentcloud/hy-video-v1.5",
  "prompt": "A lion walking through cinematic mist, slow motion, golden light",
  "aspect_ratio": "16:9",
  "callback_url": "https://example.com/hooks/video?token=unguessable"
}'

Create response

{
  "task_id": "9dcf8ab2-c4b7-42be-a516-597223897b99",
  "created": 1750000000
}

task_id is rewritten to the platform task ID.

Task response

curl "https://api.modelrover.ai/v1/tasks/9dcf8ab2-c4b7-42be-a516-597223897b99" \
-H "Authorization: Bearer $API_KEY"

{
  "task_id": "9dcf8ab2-c4b7-42be-a516-597223897b99",
  "status": "succeeded",
  "videos": [{ "url": "https://cdn.tencentcloud.com/..." }],
  "usage": { "output_seconds": 5 }
}

Task statuses are queued, running, succeeded, and failed. Video results are retained for 48 hours—download and transfer them immediately after completion.

Current limitations

hy-video-v1.5 produces 5-second, 720p output; the duration and resolution fields have no effect.
For image-to-video, the aspect ratio is determined by the input image; aspect_ratio is ignored.
Client routing and paid service-tier selections are rejected.

See Tasks and callbacks for polling, callback delivery, and task lifecycle.