Skip to main content
POST
  • sora-2-vvip video generation model
  • Supports text-to-video, image-to-video, and character reference
  • Async task management, query results by task ID
Important Change: For better performance and cost control, we no longer support passing base64 image data directly in image_urls. Please use the Upload 이미지 API first to upload images and get URLs, then call this endpoint.

Authorizations

string
필수
All endpoints require Bearer Token authenticationGet your API Key from the API Key Management PageAdd to request header:

Body

string
기본값:"sora-2-vvip"
필수
Video generation model nameSupported model: sora-2-vvip - VVIP version with fixed duration tiersExample: "sora-2-vvip"
string
필수
Text description for video generation💡 @Character Reference Feature:You can use @username format in the prompt to reference previously created characters and include them in the generated video.
  • The character’s username can be obtained via the Query Character API
  • Supports referencing multiple characters at once
Example: "A cat and a dog driving a car together @duksvfkf.cruisingki @zdqwahgj.baronbarki"
integer
Video duration in secondsSupports only 4, 8, or 12 secondsExample: 4
string
Video aspect ratioSupported formats:
  • 16:9 (Landscape)
  • 9:16 (Portrait)
string[]
Array of reference image URLs for image-to-video generation⚠️ URL format only (base64 no longer supported)
  • Publicly accessible image URL (http:// or https://)
  • You can use the Upload 이미지 API to upload local images and get URLs
boolean
Whether to generate a video thumbnail
  • true - Generate a video thumbnail for preview and display
If you don’t need a thumbnail, you can omit this parameter
integer
기본값:"1"
Number of different variants to generateRange: 1-4Default: 1
boolean
기본값:"false"
Whether to add a watermark to the generated video
  • false - No watermark
  • true - Add Sora official watermark to the video
Default: false
boolean
기본값:"false"
Whether to enable privacy mode
  • true - Video will not be published and cannot be remixed (re-edited)
  • false - Video can be published and remixed
Default: false
string
Video styleSupported values:
  • thanksgiving - Thanksgiving style
  • comic - Comic style
  • news - News style
  • selfie - Selfie style
  • nostalgic - Nostalgic/Retro style
  • anime - Anime style
Example: "anime"
boolean
Whether to use storyboard for finer control over video generation details
  • true - Enable storyboard feature
  • false - Do not use storyboard
Example: true
string
Reference video URL for character extractionUsed to reuse characters from a reference video in the generated videoExample: "https://filesystem.site/cdn/20251030/javYrU4etHVFDqg8by7mViTWHlMOZy.mp4"⚠️ Note: When using this parameter, the generated video duration will be reduced by 1 second
string
Character appearance timestampsSpecifies the time range in the reference video where the character appears, only supports 2-second segmentsFormat: "start_second,end_second"Example: "1,3" means the character from second 1 to second 3 in the reference video⚠️ Note: When using this parameter, the generated video duration will be reduced by 1 second
boolean
기본값:"false"
Automatically create a character from the generated video after completion
  • true - Automatically create character
  • false - Do not create character
Default: false
string
Create a character from an already generated task IDReference a previously generated video task ID to extract characters from that videoExample: "video_01K8SGYNNNVBQTXNR4MM964S7K"

응답

string
Unique task identifier for status queries
string
Object type, always generation.task
string
모델 name used
string
Task status
  • queued - Queued for processing
  • in_progress - Processing
  • completed - Successfully completed
  • failed - Failed
integer
Task progress percentage (0-100)
integer
Task creation timestamp (Unix timestamp)
object
Task metadata