Skip to main content
POST
Veo 3.1 Fast First/Last Frame Video Generation
Generate a video by specifying first-frame and last-frame images together with a text prompt. The model interpolates between the two frames to generate coherent motion content. Uses the Google Veo 3.1 Fast model (veo-3.1-fast-generate-001) for faster generation.
This is an asynchronous API and only returns the asynchronous task’s task_id. You should use this task_id to request the Get Task Result API to retrieve the generated result.

Request Headers

string
required
Enum value: application/json
string
required
Bearer authentication format: Bearer {{API key}}.

Request Body

integer
Random seed for deterministic generation. The same seed and parameters will produce the same result.Value range: [0, 4294967295]
string
required
First-frame image, supporting image URLs or Base64 encoding. Supports JPEG and PNG formats, up to 20MB.
string
required
Text prompt describing the video content and the motion process between the first and last frames.
string
required
Last-frame image, supporting image URLs or Base64 encoding. Supports JPEG and PNG formats, up to 20MB.
string
default:"720p"
Output video resolution.Allowed values: 720p, 1080p
string
default:"16:9"
Video aspect ratio. 16:9 is landscape, and 9:16 is portrait.Allowed values: 16:9, 9:16
integer
default:1
Number of videos to generate, ranging from 1 to 4.Value range: [1, 4]
boolean
default:true
Whether to enhance the input prompt to improve generation quality.
boolean
default:false
Whether to generate an audio track.
string
Negative prompt describing content you do not want to appear in the video.
integer
default:8
Video duration in seconds. Options: 4, 6, or 8 seconds.Allowed values: 4, 6, 8
string
default:"allow_adult"
Person generation control. allow_adult: allow generating adults; disallow: prohibit generating people.Allowed values: allow_adult, disallow

Response Information

string
Use task_id to request the Get Task Result API to retrieve the generated output.