Skip to main content
POST
/
generate
/
k
/
veo3-reference-to-video
Veo 3.1 Reference Image to Video
curl --request POST \
  --url https://api.sinancode.com/v1/generate/k/veo3-reference-to-video \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "imageUrls": [
    "https://example.com/reference1.jpg"
  ],
  "prompt": "generate a video based on the reference images",
  "enableTranslation": true,
  "seeds": 12345,
  "watermark": ""
}
'
{
  "task_id": "01234567-89ab-cdef-0123-456789abcdef"
}
Generate style-consistent videos from reference images using Google DeepMind’s Veo 3.1 model. Key Features:
  • Multi-Image Reference - Support 1-3 reference images to control character, object, or style
  • Style Transfer - Apply reference image style to generated videos
  • Character Consistency - Ideal for creating multi-shot content with consistent characters

Authorizations

​
Authorization
string
header
required

Please use your API Token for authentication

Body

application/json

Task parameters

​
imageUrls
string[]
required

Image URL list (required) - Reference image URLs (1-3)

Required array length: 1 - 3 elements
Example:
["https://example.com/reference1.jpg"]
​
prompt
string
required

Prompt (required) - Text description for video generation

Minimum string length: 1
Example:

"generate a video based on the reference images"

​
enableTranslation
boolean
default:true

Enable translation (optional) - Enable translation, default true

Example:

true

​
seeds
integer

Random seed (optional) - Range 10000-99999

Required range: 10000 <= x <= 99999
Example:

12345

​
watermark
string

Watermark (optional) - Watermark identifier

Example:

""

Response

OK

​
task_id
string
required

Task ID

Example:

"01234567-89ab-cdef-0123-456789abcdef"