Skip to main content
POST
/
v3
/
async
/
vidu-q2-turbo-startend2video
VIDU Q2 Turbo Start-End Frame to Video
curl --request POST \
  --url https://api.novita.ai/v3/async/vidu-q2-turbo-startend2video \
  --header 'Authorization: <authorization>' \
  --header 'Content-Type: <content-type>' \
  --data '
{
  "bgm": true,
  "seed": 123,
  "images": [
    {}
  ],
  "is_rec": true,
  "prompt": "<string>",
  "wm_url": "<string>",
  "payload": "<string>",
  "duration": 123,
  "off_peak": true,
  "meta_data": "<string>",
  "watermark": true,
  "resolution": "<string>",
  "wm_position": 123,
  "movement_amplitude": "<string>"
}
'
{
  "task_id": "<string>",
  "provider_request_id": "<string>"
}

Documentation Index

Fetch the complete documentation index at: https://novita.ai/docs/llms.txt

Use this file to discover all available pages before exploring further.

VIDU Q2 Turbo start-end frame to video API, Turbo version balances generation speed and video quality. Quickly generates coherent video content from start and end frame images.
This is an asynchronous API; only the task_id will be returned. You should use the task_id to request the Task Result API to retrieve the video generation results.

​
Request Headers

​
Content-Type
string
required
Supports: application/json
​
Authorization
string
required
Bearer authentication format, for example: Bearer {{API Key}}.

​
Request Body

​
bgm
boolean
default:false
Whether to add background music. true: automatically select suitable music from preset BGM library; false: no BGM. BGM duration adapts to video length automatically.
​
seed
integer
Random seed. Default or 0 uses random number. Manual setting uses specified seed.
​
images
array
required
Images array, first image is start frame, second image is end frame. Supports two images. Note: 1) Resolution ratio of start/end frames should be between 0.8-1.25; 2) Supports Base64 encoding or image URL; 3) Supports png, jpeg, jpg, webp formats; 4) Max size 50M per image; 5) Base64 must include proper content type, e.g., data:image/png;base64,Array length: 2 - 2
​
is_rec
boolean
default:false
Whether to use recommended prompt. true: system automatically recommends and uses prompt (consumes 10 extra credits); false: use input prompt
​
prompt
string
Text prompt describing the desired video content. Max 2000 characters. Note: If is_rec is enabled, this parameter will be ignoredLength limit: 0 - 2000
​
wm_url
string
Watermark content image URL. If not provided, uses default watermark: ‘Generated by AI’
​
payload
string
Pass-through parameter, no processing, data transmission only. Max 1048576 characters.Length limit: 0 - 1048576
​
duration
integer
default:5
Video duration in seconds, supports 1-8 secondsOptional values: 1, 2, 3, 4, 5, 6, 7, 8
​
off_peak
boolean
default:false
Off-peak mode. true: off-peak generation (lower credits, completed within 48 hours); false: immediate generation. Note: Off-peak tasks can be manually cancelled and credits refunded if not completed.
​
meta_data
string
Metadata identifier in JSON string format. Pass-through field. You can define custom format or use example format. When empty, uses Vidu-generated metadata.
​
watermark
boolean
default:false
Whether to add watermark. Default: no watermark. Query watermarked video via watermarked_url parameter.
​
resolution
string
default:"720p"
Output video resolution. Default is 720p.Optional values: 540p, 720p, 1080p
​
wm_position
integer
default:3
Watermark position: 1=top-left, 2=top-right, 3=bottom-right (default), 4=bottom-leftOptional values: 1, 2, 3, 4
​
movement_amplitude
string
default:"auto"
Movement amplitude, controls the intensity of object movementOptional values: auto, small, medium, large

​
Response

​
task_id
string
required
Use the task_id to request the Task Result API to retrieve the generated outputs.
​
provider_request_id
string
Provider request ID (optional)