mingyanmingyan
wechatwechat
mingyan

Vidu - Create Task

Create video generation tasks using Vidu models, supporting image-to-video and first/last frame guidance

POST/kyyReactApiServer/v1/vidu/videos

Vidu Create Video Task

Create video generation tasks using Vidu's AI models, supporting text-to-video, image-to-video, and first/last frame guidance modes.

Authentication

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

Base URL

https://zcbservice.aizfw.cn/kyyReactApiServer
baseUrl is the shared prefix for all public APIs. The api field in the current page frontmatter shows the full endpoint. Use this baseUrl as the common prefix when reading or composing request paths.

Model Introduction

Supported models:
  • viduq3-pro: Efficiently generates high-quality audio-visual content, making videos more vivid, vivid, and three-dimensional, with better results
  • viduq3-turbo: Compared to viduq3-pro, faster generation speed

Request Parameters

modelstringrequired
Model name
Available values:
  • viduq3-pro - Vidu Q3 Pro, better quality
  • viduq3-turbo - Vidu Q3 Turbo, faster speed
promptstringrequired
Video generation prompt describing the desired video content
Example: "A cat dancing in the rain, cinematic style"
durationnumber
Video duration in seconds, default value depends on the model
  • viduq3-pro, viduq3-turbo: default 5 seconds
  • Available range: 1-16 seconds
aspect_ratiostring
Output video aspect ratio, default is 16:9
Supported values:
  • 16:9 - Landscape
  • 9:16 - Portrait
  • 1:1 - Square
  • 4:3 - Standard
  • 3:4 - Portrait standard
resolutionstring
Output resolution, default is 720p
Supported values:
  • 540p - Standard definition
  • 720p - High definition (default)
  • 1080p - Full HD

Image-to-Video Mode

first_imagestring
First frame image URL (image-to-video scenario)
Usage:
  • Use alone: image-to-video scenario
  • Use with last_image: first/last frame scenario
Format requirements: jpeg, png, webp Value type: Public URL
last_imagestring
Last frame image URL (first/last frame scenario, must be used with first_image)
Format requirements: jpeg, png, webp Value type: Public URL
Must be used together with first_image

Response Parameters

idstring
Unique identifier for the video generation task, used for subsequent status queries
objectstring
Object type, fixed as video
createdinteger
Task creation timestamp
modelstring
Model name used
statusstring
Task status:
  • queued - Queued
  • processing - Processing
  • completed - Completed
  • failed - Failed
errorstring
Error message, returned when status is failed

Use Cases

Text-to-Video

Generate videos using only prompt, no image guidance needed

Image-to-Video

Use first_image as the first frame to generate coherent video

First/Last Frame Guidance

Use both first_image and last_image for precise control of start and end

Vidu Features

Flexible Duration

Supports adjustable video duration from 1-16 seconds

Multiple Aspect Ratios

Supports 5 common video aspect ratios

Multiple Resolutions

Supports 540p, 720p, 1080p resolutions

Dual Model Selection

Pro version for better quality, Turbo version for faster speed
Best Practices:
  1. Video generation is asynchronous, use the returned id to query task status and results
  2. Clearly describe the desired visual effects and style in the prompt
  3. Choose the appropriate model:
    • viduq3-pro: Efficiently generates high-quality audio-visual content, better results
    • viduq3-turbo: Faster generation speed
  4. Select appropriate duration based on needs (1-16 seconds)
  5. Image URLs must be publicly accessible
Scenario Selection Recommendations:
  • Text-to-Video: Fully AI-created, suitable for creative exploration
  • Image-to-Video: Generated based on first frame image, suitable for scenarios with clear starting visuals
  • First/Last Frame Guidance: Precise control of start and end, suitable for scenarios requiring specific transition effects