Sora Videos

The Sora series currently provides two integration methods: the OpenAI-compatible /v1/videos route, and the unified video /v1/video/create and /v1/video/query.

OpenAI Format

Use POST /v1/videos to submit and GET /v1/videos/{task_id} to query; reference images are expressed via input_reference.

Sora Video Generation

POST /v1/videos, submit Sora tasks in OpenAI style.

Sora Task Query

GET /v1/videos/{task_id}, query the OpenAI-style video object.

Current Models (OpenAI Format)

  • sora-2

Unified Video

Use POST /v1/video/create to create tasks and GET /v1/video/query?id= to query; reference images are passed via the images array.

Create Video

POST /v1/video/create, submit Sora tasks in the unified video format.

Query Task

GET /v1/video/query, query status and results by task id.

Common Models (Unified Video)

  • sora-2
  • sora-2-pro (supports longer duration)

Comparison

ItemOpenAI FormatUnified Video
CreatePOST /v1/videosPOST /v1/video/create
QueryGET /v1/videos/{task_id}GET /v1/video/query?id=
Reference Imageinput_referenceimages
Frame Sizesize, metadatasize, orientation
Durationsecondsduration
Proprietary Optionsmetadatawatermark, private, style, storyboard
In Apifox, character_url and character_timestamps belong to the Sora character creation workflow and correspond to a separate route:
POST /sora/v1/characters
They are not documented as primary fields for POST /v1/videos or POST /v1/video/create; see the explanation in Sora Video Generation.