OpenAI Sora 2 (reverse) Generate Video
Video
OpenAI Sora 2 (reverse) Generate Video
POST
OpenAI Sora 2 (reverse) Generate Video
Sora 2 (reverse) is a state-of-the-art video + audio generator. It improves on the original Sora with more accurate physics, clearer realism, synchronized audio, stronger controllability, and a broader range of styles.
Request Headers
Enum value:
application/jsonBearer authentication format: Bearer {{API Key}}.
Request Body
The positive text prompt for video generation.
The input image only supports URL format. Supported image formats include .jpg, .jpeg, and .png.
Video duration (seconds)Enum values:
10,15Width x height. The specific values are ignored: width > height indicates landscape, and width < height indicates portrait.Enum values:
1280x720- Standard-definition landscape (actual output resolution is approximately 480P; marked as 480P in the pricing column)720x1280- Standard-definition portrait (actual output resolution is approximately 480P; marked as 480P in the pricing column)
Whether a watermark is required. If omitted, the default is no watermark.
The video link required to create a character. Note that no real person may appear in the video, otherwise it will fail.
The time range in seconds during which the video character appears, in the format {start},{end}. Note that the end-start range must be 1–3 seconds.Example value:
1,3Response Information
The task_id of the asynchronous task. You should use this task_id to request the Get Task Result API to obtain the generation result.