Sora-2（Official Format）

Sora is OpenAI’s latest video generation model, capable of generating detailed and dynamic videos based on text or images.
The model is based on multimodal diffusion research and possesses a deep understanding of 3D space, motion, and scene continuity.

Usage Process

1. Create video
   Use the Create video endpoint, set the prompt, model, duration, resolution, and reference image. After submission, a task ID will be returned.

2. Check progress
   Use the Retrieve video endpoint with the task ID to check the status.

3. Get video
   When "status": "completed", use the Retrieve video content endpoint to obtain the video link.

Remix Feature

sora2 supports Remix video, allowing secondary creation based on an existing video’s ID and a new prompt, enabling “existing video + new intent” regeneration.

Tip: Since the Sora-2 image-to-video function has very strict requirements on the pixel of the input image, the uploaded images are only limited to the following sizes: 1280x720, 720x1280, 1024x1792, 1792x1024

Pricing

sora2 supports Remix video, allowing secondary creation based on an existing video ID and new prompt, enabling “existing video + new intent” regeneration.
Fill in the existing video_id in Params, and enter the prompt in Body via prompt.

Price: sora2 720x1280 or 1280x720 0.1PTC/second
Price: sora-2-pro 720x1280 or 1280x720 0.3 PTC/second
Price: sora-2-pro 1024x1792 or 1792x1024 0.5 PTC/second

Enter the task ID to check the generation progress
Price: 0 PTC

Delete a generated video by video_id
Price: 0 PTC