Text to Video

Endpoint

POST http://103.36.63.156:3000/v1/videos
Authorization: Bearer {api_key}
Content-Type: application/json

Request Example

curl -X POST "http://103.36.63.156:3000/v1/videos" \
  -H "Authorization: Bearer sk-xxxxxxxx" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "minimax_h3",
    "prompt": "Cyberpunk city at night, neon lights reflecting on rain-soaked streets, slow camera push-in",
    "size": "1920x1080",
    "duration": 6,
    "aspect_ratio": "16:9"
  }'

Basic Parameters

Parameter	Type	Required	Description
`model`	string	Yes	Fixed to `minimax_h3`
`prompt`	string	Yes	Video content prompt
`size`	string	No	Output size, formatted as `widthxheight`
`aspect_ratio`	string	No	Aspect ratio
`duration`	number	No	Video duration (seconds), 5–15
`megapixels`	number	No	Target megapixel count
`callback_url`	string	No	Callback URL invoked on task completion

Framing and Clarity

Framing and clarity can be controlled in three ways: size, aspect_ratio, and megapixels.

Method	Description
`size`	Specify pixel dimensions directly; highest priority
`aspect_ratio`	Specify a ratio; the server derives a suitable size
`megapixels`	Specify the megapixel count, used together with a ratio

Supported Aspect Ratios

adaptive / 21:9 / 16:9 / 4:3 / 1:1 / 3:4 / 9:16

Standard Pixel Sizes

Ratio	Size Examples
16:9	1280x720 / 1920x1080
9:16	720x1280 / 1080x1920
1:1	1024x1024
4:3	1024x768
3:4	768x1024
21:9	1680x720

When neither size nor aspect_ratio is provided, the server uses adaptive mode and determines the framing from the prompt itself.

Response

{
  "task_id": "xxxxxxxxxxxxxxxx",
  "status": "queued"
}

Once you have the task_id, poll the result via the Query Task and Download endpoint.