Skip to main content
POST
Generate a video

Authorizations

Authorization
string
header
required

Bearer token used for request authentication. Alternatively, you can send the API key in the x-api-key header. Note: A payment method is required to use the API.

Body

application/json

JSON request body for the Video Generation API.

model
enum<string>
required

The model to use.

Available options:
minimax/minimax-h3,
bytedance/seedance-2.5,
google/veo-3.1,
google/veo-3.1-fast,
google/veo-3.1-lite
prompt
string
required

The text description of the video.

duration
integer

Video length in seconds. When omitted, the model default applies.

resolution
enum<string>

Output resolution. When omitted, the model default applies.

Available options:
480p,
720p,
768p,
1080p,
2K
aspect_ratio
enum<string>

Aspect ratio. adaptive lets the model follow the input. When omitted, the model default applies.

Available options:
16:9,
4:3,
1:1,
3:4,
9:16,
21:9,
adaptive
frame_images
object[]

Images to use as the first and last frames of the video.

input_references
object[]

Reference images that guide generation without locking specific frames.

generate_audio
boolean
default:true

Whether to generate audio along with the video.

seed
integer

Random seed, for more reproducible results.

extra_params
object

Vendor-specific parameters, passed through to the upstream model without validation. Keys that duplicate a unified parameter, or that control upstream storage, notification, sample count, or platform behavior, are rejected.

Response

Successful video generation response. The task starts in queued.

id
string
required

Unique task id, used to download the video.

model
string
required

The model used for this task.

status
enum<string>
required

Task state.

Available options:
queued
created
integer
required

Unix timestamp in seconds when the task was created.