Skip to main content
POST
/
v1
/
veo
/
videos
curl --request POST \
  --url https://api.globalaiopc.com/veo/videos \
  --header 'Authorization: Bearer {{key}}' \
  --header 'Content-Type: application/json' \
  --data '{
    "model": "veo3.1-fast",
    "prompt": "A cat dancing in the rain, cinematic style",
    "aspect_ratio": "16:9",
    "seconds": 8,
    "input_reference": [
      "https://example.com/reference1.jpg",
      "https://example.com/reference2.jpg"
    ]
  }'
{
  "id": "vgen_abc123def456",
  "object": "video.generation",
  "created": 1761635478,
  "model": "veo3.1-fast",
  "status": "queued"
}

VEO Create Video Task

Create high-quality video generation tasks using VEO’s advanced AI models with support for two image guidance modes.

Authentication

All requests require a Bearer token in the request header:
Authorization: Bearer {{key}}

Request Parameters

model
string
Model name, defaults to veo3.1-fastAvailable options:
  • veo3.1-fast - Fast generation model
prompt
string
required
Video generation prompt describing the desired video contentExample: "A cat dancing in the rain, cinematic style"
size
string
Video dimensions, defaults to 1280x720Available options:
  • 1280x720 - Landscape
  • 720x1280 - Portrait
Use either this or aspect_ratio, not both
aspect_ratio
string
Video aspect ratioAvailable options:
  • 16:9 - Landscape
  • 9:16 - Portrait
Use either this or size, not both
duration
integer
Video duration in seconds, supports 8, defaults to 8
Use either this or seconds, not both
seconds
integer
Video duration in seconds, supports 8, defaults to 8
Use either this or duration, not both

Reference Image Mode

input_reference
array
Reference image array to guide video generation, supports up to 3 images
Mutually exclusive with first/last frame mode, choose either this or image_urls
image_urls
array
Reference image URL array to guide video generation, supports up to 3 images
Mutually exclusive with first/last frame mode, choose either this or input_reference

First/Last Frame Mode

first_image
string
First frame image URL to specify the video’s starting frame
Mutually exclusive with reference image mode
last_image
string
Last frame image URL to specify the video’s ending frame
Requires first_image to be provided, mutually exclusive with reference image mode

Response Parameters

id
string
Unique identifier for the video generation task, used for subsequent status queries
object
string
Object type, always video.generation
created
integer
Task creation timestamp
model
string
Model name used
status
string
Task status:
  • queued - Queued
  • processing - Processing
error
string
Error message, returned when status is failed

Image Guidance Modes

VEO supports two mutually exclusive image guidance modes:

Reference Image Mode

Use input_reference or image_urls to provide up to 3 reference images for overall style and content guidance

First/Last Frame Mode

Use first_image and optional last_image to specify precise start and end frames of the video
Important: The two modes cannot be used simultaneously, choose only one

Parameter Selection Rules

Mutually Exclusive Parameters:
  • Size parameters: Choose either size or aspect_ratio
  • Duration parameters: Choose either duration or seconds
  • Reference images: Choose either input_reference or image_urls
  • Guidance modes: Reference image mode OR first/last frame mode, cannot use both

VEO Features

Fixed Duration

VEO model generates fixed 8-second videos

Dual Guidance Modes

Supports both reference image and first/last frame guidance

High Quality Output

veo3.1-fast model provides high-quality video generation

Flexible Control

First/last frame mode provides precise start and end control
Best Practices:
  1. Video generation is asynchronous, use the returned id to query task status and results
  2. Clearly describe desired visual effects and style in the prompt
  3. Choose appropriate guidance mode:
    • Reference image mode: Suitable for style guidance
    • First/last frame mode: Suitable for precise start and end control
  4. If using first/last frame mode, always provide the first frame image
Guidance Mode Selection:
  • Reference Image Mode: Use when you want overall style consistency but allow creative variation
  • First/Last Frame Mode: Use when you need precise control over video start and end frames
curl --request POST \
  --url https://api.globalaiopc.com/veo/videos \
  --header 'Authorization: Bearer {{key}}' \
  --header 'Content-Type: application/json' \
  --data '{
    "model": "veo3.1-fast",
    "prompt": "A cat dancing in the rain, cinematic style",
    "aspect_ratio": "16:9",
    "seconds": 8,
    "input_reference": [
      "https://example.com/reference1.jpg",
      "https://example.com/reference2.jpg"
    ]
  }'
{
  "id": "vgen_abc123def456",
  "object": "video.generation",
  "created": 1761635478,
  "model": "veo3.1-fast",
  "status": "queued"
}