Skip to main content

​
Introduction to Sora

Welcome to the Sora service documentation! This service integrates Sora AI, OpenAI’s revolutionary text-to-video generation model. Whether you’re looking to create stunning videos from text descriptions, generate creative content, or explore new video creation capabilities, Sora provides the ideal tools for your project.

​
Features

Sora AI offers advanced video generation capabilities to enhance your creative work:
  • Text-to-Video Generation: Create high-quality videos from text prompts using the powerful Sora-2 model. Simply describe what you want to see, and Sora will generate a video that brings your vision to life.
  • Flexible Video Formats: Generate videos in different orientations (landscape or portrait) and sizes (small or large) to match your specific needs.
  • Customizable Duration: Create videos with durations of 10 or 15 seconds, giving you control over the length of your content.
  • Image-to-Video: Use reference images to guide your video generation, creating more precise and contextually relevant content.
  • Watermark Control: Choose whether to include or remove watermarks from your generated videos.
  • Task Management: Track and monitor the status of your video generation tasks to ensure smooth processing.

​
How to Use Sora

To get started with Sora, follow these simple steps:
  1. Create a Video Generation Task: Start by creating a task with your text prompt and desired parameters.
  2. Track Task Status: Monitor the task’s progress to know when your video is ready for download.
  3. Download and Use: Once your task is complete, you can download the generated video for use in your project.

​
Available Endpoints

Here are the API endpoints for interacting with Sora:

​
Video Generation API

  • Generate Video with Sora-2 Model
    POST /sora/video/generate
    Create a video using the Sora-2 model. Provide a text prompt describing the video you want to generate, along with optional parameters like orientation, size, duration, and reference images.
  • Get Task Result
    GET /sora/task/result
    Retrieve the current status and result of your video generation task. Check if your video is ready for download or if there are any errors.

​
Explore the API Documentation

For detailed documentation on how to use these endpoints, click the links below:

​
Key Parameters

When generating videos with Sora, you can customize several important parameters:
  • Prompt: A detailed text description of the video you want to generate (max 5000 characters)
  • Model: Currently supports sora-2 (default)
  • Orientation: Choose between landscape (default) or portrait
  • Size: Select small (default) or large for different video dimensions
  • Duration: Set to 10 (default) or 15 seconds
  • Images: Optional array of reference image URLs to guide video generation
  • Remove Watermark: Boolean flag to control watermark presence (default: false)

​
Best Practices

To get the best results from Sora:
  1. Write Detailed Prompts: Be specific and descriptive in your text prompts to get more accurate results
  2. Use Reference Images: When available, provide reference images to guide the video generation
  3. Choose Appropriate Settings: Select the right orientation and size for your intended use case
  4. Monitor Task Status: Regularly check your task status to ensure successful completion
  5. Handle Errors Gracefully: Implement proper error handling for failed tasks
For more details, visit the full Sora API Documentation.