Authentication
authorization string required
All APIs require authentication via Bearer Token.
Get API Key:
Visit API Key Management Page to get your API Key.
Usage:
Add to request header:
Authorization: Bearer YOUR_API_KEY
Parameters
model string required
Model ID used for this request
Value: viduq3_turbo_start_end2video
images array<string> required
Two-image input for start-end generation. The first image is treated as the start frame, the second image is treated as the end frame
Note 1: Start/end image resolutions should be close. Start resolution / end resolution must be between 0.8 and 1.25, and aspect ratio must be less than 1:4 or 4:1
Note 2: Supports image Base64 or image URL (must be accessible)
Note 3: Supported formats: png, jpeg, jpg, webp
Note 4: Max image size is 50MB
Note 5: HTTP POST body must be under 20MB, and encoding must include content type, for example data:image/png;base64,{base64_encode}
prompt string
Text prompt describing the video to generate
Limit: Up to 5000 characters
is_rec boolean
Whether to use a recommended prompt
true: System auto-generates one recommended prompt and uses it for generation
false: Uses the provided prompt
Note: Enabling recommended prompts consumes an extra 10 credits per task
Options: true, false
Default: false
duration integer
Video duration in seconds
Options: 1 - 16
Default: 5
seed integer
Random seed. If omitted or set to 0, a random seed is used automatically
resolution string
Output resolution
Options: 540p, 720p, 1080p
Default: 720p
audio boolean
Whether to enable direct audio-video generation
false: Outputs a silent video
true: Outputs a video with speech and sound effects
Options: true, false
Default: true
payload string
Pass-through parameter for custom payload data
Limit: Up to 1048576 characters
off_peak boolean
Note: Temporarily unavailable
Off-peak generation mode
true: Off-peak generation
false: Immediate generation
Note 1: Off-peak mode consumes fewer credits
Note 2: Tasks are processed within 48 hours, unfinished tasks are auto-canceled with credit refund
Note 3: You can also cancel off-peak tasks manually
Options: true, false
Default: false
watermark boolean
Whether to add a watermark
true: Add watermark
false: No watermark
Note 1: Current watermark content is fixed and AI-generated, disabled by default
Note 2: You can retrieve watermarked content via watermarked_url
Options: true, false
Default: false
wm_position integer
Watermark position in the frame
1: Top-left
2: Top-right
3: Bottom-right
4: Bottom-left
Default: 3
wm_url string
Watermark image URL. If omitted, the default AI-generated watermark is used
meta_data string
Metadata identifier in JSON string format, passed through without modification
{
"Label": "your_label",
"ContentProducer": "yourcontentproducer",
"ContentPropagator": "your_content_propagator",
"ProduceID": "yourproductid",
"PropagateID": "your_propagate_id",
"ReservedCode1": "yourreservedcode1",
"ReservedCode2": "your_reserved_code2"
}
If empty, default metadata generated by vidu is used
callback_url string
Callback URL. Set callback_url when creating the task, request method is POST
When task status changes, Vidu will send callback payloads matching the task-query API response structure
Polling
Since result generation takes time, you need to poll the task status after creating the task.
The initial response only returns information such as the task ID and initial status. The final result must be obtained by polling the task status endpoint using the task ID.
See the examples on the right for polling requests and responses.
Response Format
error object
Error information, only present when status is failed
code
integerError code
error_message
stringDetailed error message
output array
Generation result, only present when status is completed
content
arrayGenerated content list
type
stringResource type
Value:
videourl
stringGenerated content URL
usage object
Usage statistics, only present when status is completed
cost
stringTotal cost in USD
discount
numberDiscount amount
metadata object
Metadata information