# Seedance 2.0 - Core.Today AI API

> ByteDance's next-generation multimodal video model with native synchronized audio. Combines up to 9 reference images, 3 videos, and 3 audio files in a single generation for character-consistent, lip-synced video creation, editing, and extension.

- **Provider**: ByteDance
- **Model ID**: bytedance/seedance-2.0
- **Category**: Video Generation
- **Credits**: 2090 per 5s 720p video — billed per second (480p 186/s, 720p 419/s, 1080p 1047/s, 4K 2327/s; with reference videos 480p 233/s, 720p 512/s, 1080p 1280/s, 4K 2908/s)
- **Speed**: Medium
- **Quality**: Ultra

## Features
- Native audio generation — dialogue, sound effects, and background music synced with visuals
- Multimodal reference inputs: up to 9 images, 3 videos, 3 audio files in one pass
- Character consistency across shots using reference images
- Video editing and extension from a reference video
- Up to 4K (10-bit H.265) resolution and 1-15 second duration

## Use Cases
- Dialogue scenes with lip-synced speech (quote dialogue in the prompt)
- Outfit-change and product showcase videos with reference images
- Music-synced short-form content using reference audio
- Extending or editing existing footage while preserving motion
- Multi-shot narratives with consistent characters

## 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 - Text prompt (max 4000 chars). Use double quotes for dialogue; reference inputs as [Image1], [Video1], [Audio1]

### Optional
- **image**: string - First frame image for image-to-video (cannot combine with reference images)
- **last_frame_image**: string - Last frame image (requires a first frame image)
- **reference_images**: array (default: []) - Up to 9 reference images for character/style consistency
- **reference_videos**: array (default: []) - Up to 3 reference videos (total ≤15s) for motion transfer and editing. Raises per-second price ~25%
- **reference_audios**: array (default: []) - Up to 3 reference audio files (total ≤15s) for audio-driven generation and lip-sync
- **duration**: integer (default: 5) - Video duration in seconds (1-15). Billed per second of output
- **resolution**: string (default: 720p) - Video resolution. 4K outputs 10-bit H.265/HEVC
  Options: 480p, 720p, 1080p, 4k
- **aspect_ratio**: string (default: 16:9) - Aspect ratio. 'adaptive' lets the model choose based on inputs
  Options: 16:9, 4:3, 1:1, 3:4, 9:16, 21:9, 9:21, adaptive
- **generate_audio**: boolean (default: true) - Generate synchronized audio (dialogue, SFX, background music)
- **seed**: integer - Random seed for reproducible generation

## Examples

### Cinematic Scene with Audio
Text-to-video with synchronized ambient audio
```json
{
  "model": "bytedance/seedance-2.0",
  "input": {
    "prompt": "A cozy cabin in a snowy forest at night, warm light glowing from the windows, gentle snowfall, camera slowly pushing in through the trees",
    "duration": 7,
    "resolution": "720p",
    "aspect_ratio": "16:9",
    "generate_audio": true
  }
}
```

### Dialogue with Lip-Sync
Spoken dialogue generated with matching lip movements
```json
{
  "model": "bytedance/seedance-2.0",
  "input": {
    "prompt": "An old fisherman on a pier at dawn turns to the camera and says: \"The sea gives, and the sea takes.\" Waves lap against the wooden posts",
    "duration": 8,
    "resolution": "720p",
    "aspect_ratio": "16:9",
    "generate_audio": true
  }
}
```

## 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
- Billing is per second of output — start with 5 seconds and 720p while iterating, then scale up
- Put spoken dialogue in double quotes ("...") to get matching lip movements and voice
- Label reference inputs in the prompt: 'The character from [Image1] performs the dance from [Video1]'
- Reference videos raise the per-second price ~25% — use them only when you need motion transfer or editing
- 4K costs 5.5x more per second than 720p — reserve it for final renders

## Documentation
https://replicate.com/bytedance/seedance-2.0