# Seedance 2.0 API via PiAPI

Base URL: https://api.piapi.ai
Auth: X-API-Key header (get your key at https://piapi.ai/workspace/key)

Documentation: https://piapi.ai/docs/seedance-api/seedance-2

## Available Models (price per second of generated video)

- seedance-2: 480p $0.10/s, 720p $0.20/s, 1080p $0.50/s (Pro quality, recommended)
- seedance-2-fast: 480p $0.08/s, 720p $0.16/s (fastest)
- seedance-2-mini: 480p $0.07/s, 720p $0.14/s (~2x faster than fast, lowest cost)
- seedance-2-less-restriction: 480p $0.11/s, 720p $0.22/s, 1080p $0.55/s
- seedance-2-fast-less-restriction: 480p $0.088/s, 720p $0.176/s
- seedance-2-mini-less-restriction: 480p $0.077/s, 720p $0.154/s

Cost formula: credits = unit_price x output_duration + (unit_price / 2) x total_input_video_duration

## Text to Video

POST /api/v1/task
{
  "model": "seedance",
  "task_type": "seedance-2",
  "input": {
    "prompt": "A woman sings and strums her guitar",
    "mode": "text_to_video",
    "duration": 5,
    "aspect_ratio": "16:9",
    "resolution": "720p"
  }
}

## Generation Modes (input.mode)

- text_to_video: prompt only
- first_last_frames: pass 1-2 image_urls as start/end frames (aspect ratio follows the image)
- omni_reference: up to 12 mixed references — image_urls (max 9), video_urls (max 9, mp4/mov),
  audio_urls (mp3/wav, up to 15s total); reference them in the prompt as @image1, @video1, @audio1

## Input Parameters

- prompt (string, required, max 4000 chars)
- mode (string, required): text_to_video | first_last_frames | omni_reference
- duration (int, 4-15, default 5)
- aspect_ratio (string): 21:9 | 16:9 | 4:3 | 1:1 | 3:4 | 9:16 | auto (default 16:9)
- resolution (string): 480p | 720p | 1080p (default 480p; mini defaults to 720p; 1080p only on seedance-2 tiers)
- image_urls / video_urls / audio_urls (arrays, see modes above)
- audio (bool, default true): generate audio track

Face / real-person references: upload images to the Private Asset Library
(https://piapi.ai/docs/seedance-api/private-asset), wait for status=Active, then use
asset://<asset_id> in the -less-restriction task types.

## Get Task Result

GET /api/v1/task/{task_id}

Status: "pending" -> "processing" -> "completed" or "failed"
Video URL: response.data.output.video

## Quick Start

1. Get API key: https://piapi.ai/workspace/key
2. Top up: https://piapi.ai/workspace/billing
3. POST /api/v1/task
4. Poll GET /api/v1/task/{task_id} until "completed"
5. Get video from response.data.output.video