Skip to main content
POST
/
v1
/
videos
/
{task_id}
/
remix
curl --request POST \
  --url https://api.apimart.ai/v1/videos/{task_id}/remix \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '{
    "model": "veo3.1-fast",
    "prompt": "The cat continues running on the grass, butterflies fly into the distance"
  }'
{
  "code": 200,
  "data": [
    {
      "status": "submitted",
      "task_id": "task_01JQXYZ9999NEWEXTENDID"
    }
  ]
}
curl --request POST \
  --url https://api.apimart.ai/v1/videos/{task_id}/remix \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '{
    "model": "veo3.1-fast",
    "prompt": "The cat continues running on the grass, butterflies fly into the distance"
  }'
{
  "code": 200,
  "data": [
    {
      "status": "submitted",
      "task_id": "task_01JQXYZ9999NEWEXTENDID"
    }
  ]
}

Path Parameters

task_id
string
required
Original video task IDThis is the task_id returned from the video generation API. The original video task status must be success

Authorizations

Authorization
string
required
All API endpoints require Bearer Token authenticationGet your API Key:Visit the API Key Management Page to get your API KeyAdd it to the request header:
Authorization: Bearer YOUR_API_KEY

Body

model
string
required
Video model name, must match the model used for the original videoSupported models:
  • veo3.1-fast - Fast version
  • veo3.1-quality - High quality version
Example: "veo3.1-fast"
prompt
string
required
Continuation prompt describing the content for the extended portion

Important Notes

Model must match: The model used for extension must be the same as the one used to generate the original video. Videos generated with veo3.1-fast must also use veo3.1-fast for extension.
Only completed videos supported: The original video task status must be success.
Fixed extension duration: Automatically extends from 8 seconds to 15 seconds. No need to specify duration or aspect_ratio.
task_id in URL: This is the task_id returned from the first step of video generation (not the upstream ID).

Response

code
integer
Response status code
data
object[]
Response data array