Sora 2
OpenAI's advanced text-to-video and image-to-video generation
Sora 2
OpenAI's state-of-the-art video generation model.
Model ID
sora-2
Price
$0.10-0.15/video
Speed
2-5 minutes
Type
Video
Capabilities
| Mode | Description |
|---|---|
text-to-video | Generate video from text descriptions |
image-to-video | Animate still images into video |
Parameters
For common parameters (request_type, callback_url), see Request Format.
| Parameter | Type | Required | Default | Description |
|---|---|---|---|---|
mode | string | Yes | - | text-to-video or image-to-video |
prompt | string | Yes | - | Video description (max 10,000 characters) |
image_url | string | No | - | Starting image for image-to-video |
aspect_ratio | string | No | 16:9 | 16:9, 9:16, or 1:1 |
duration | string | No | 10 | Video duration: 10 or 15 seconds |
n_frames | integer | No | - | Number of frames (model-dependent) |
Examples
Text to Video
Image to Video
Tips
- Polling: Wait 120 seconds, then poll every 15 seconds
- Timeout: Set max wait to 10 minutes
- Longer videos take proportionally longer to generate
- Be specific about camera movements and scene dynamics