# OpenAI Sora 2 Pro - Core.Today AI API

> OpenAI's most advanced synced-audio video generation model. The premium tier of Sora 2 with higher fidelity, up to 1024p resolution, and image-to-video via an input reference frame.

- **Provider**: OpenAI
- **Model ID**: openai/sora-2-pro
- **Category**: Video Generation
- **Credits**: 2790 per 4s standard (720p) video (default) — billed per second: standard 698/s, high (1024p) 1163/s; durations 4/8/12s
- **Speed**: Slow
- **Quality**: Ultra

## Features
- OpenAI's most advanced video model with natively synced audio
- Premium tier of Sora 2 — higher fidelity and detail
- Two resolution tiers: standard (720p) and high (1024p)
- Image-to-video via input_reference first frame
- Portrait or landscape output, 4/8/12 second durations

## Use Cases
- Premium marketing and ad footage with synced dialogue and sound
- Realistic physics-driven scenes with coherent audio design
- High-resolution (1024p) hero shots for brand campaigns
- Vertical short-form content with natural speech and ambience
- Animating a key art frame into a full audio-visual clip

## API Endpoint

Base URL: https://api.core.today/v1
Create Prediction: POST /predictions
Get Status: GET /predictions/{job_id}

## Authentication
Header: X-API-Key: YOUR_API_KEY

## Input Parameters

### Required
- **prompt**: string - A text description of the video to generate

### Optional
- **seconds**: integer (default: 4) - Duration of the video in seconds
  Options: 4, 8, 12
- **resolution**: string (default: standard) - Resolution quality. Standard is 720p (1280x720 landscape, 720x1280 portrait). High is 1024p (1792x1024 landscape, 1024x1792 portrait)
  Options: standard, high
- **aspect_ratio**: string (default: portrait) - Aspect ratio of the video
  Options: portrait, landscape
- **input_reference**: string - An optional image to use as the first frame of the video. Must match the video's aspect ratio
- **openai_api_key**: string - Optional: your own OpenAI API key (you would then be charged directly by OpenAI). Platform billing here is unchanged either way

## Examples

### Character Dialogue with Synced Audio
A standard-tier portrait clip with natively generated speech and ambience
```json
{
  "model": "openai/sora-2-pro",
  "input": {
    "prompt": "Scottish Highland coo with ginger fur getting a parking ticket from a Glaswegian police officer speaking in a thick accent, parked on a double yellow line in a small Scottish town",
    "seconds": 4,
    "resolution": "standard",
    "aspect_ratio": "portrait"
  }
}
```

### High-Resolution Landscape Hero Shot
A 1024p landscape render at the high tier for final delivery
```json
{
  "model": "openai/sora-2-pro",
  "input": {
    "prompt": "Cinematic drone shot gliding over a mist-covered alpine lake at sunrise, golden light breaking through pine forests, gentle wind and distant birdsong",
    "seconds": 8,
    "resolution": "high",
    "aspect_ratio": "landscape"
  }
}
```

## Response Format
```json
{
  "job_id": "abc123",
  "status": "pending | processing | completed | failed",
  "result": "URL or data (when completed)"
}
```

## Usage Flow
1. POST /predictions with model and input -> receive job_id
2. GET /predictions/{job_id} -> poll until status is completed or failed
3. Result contains output URL(s)

## Tips
- high resolution (1024p) costs 1.67x the standard rate (750/s vs 450/s) — iterate at standard and reserve high for final renders
- Duration only accepts 4, 8, or 12 seconds; a 4s standard clip is the cheapest way to test a prompt (1800 credits)
- aspect_ratio defaults to portrait — set it to landscape explicitly for widescreen content
- input_reference must match the video's aspect ratio, so prepare the image in the target orientation first
- Supplying your own openai_api_key routes upstream charges to OpenAI directly, but does not change credit billing on this platform

## Documentation
https://replicate.com/openai/sora-2-pro