Seedance 2.0 API

The API lets you submit Seedance 2.0 video generation tasks programmatically. Generation is asynchronous: create a task, receive a task ID immediately, then get the finished video by polling the task endpoint or by receiving a webhook.

Create an API key in the dashboard and send it as a Bearer token on every request. The full key is shown only once at creation time.

Authorization: Bearer sk_live_xxxxxxxx

Submit a task first. After the task is accepted, choose one result delivery method: poll the task endpoint, or receive the terminal result through a webhook.

curl https://api.seedance2.ai/v1/videos/generations \
  -H "Authorization: Bearer sk_live_xxx" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "seedance-2-0",
    "callback_url": "https://your-domain.com/api/seedance/webhook",
    "input": {
      "prompt": "a cat surfing on a neon wave, cinematic lighting",
      "generation_type": "text-to-video",
      "duration": 5,
      "aspect_ratio": "16:9",
      "resolution": "720p",
      "generate_audio": true,
      "watermark": false,
      "web_search": false,
      "return_last_frame": false,
      "seed": -1
    }
  }'

curl https://api.seedance2.ai/v1/tasks/3f2aK9mR7xQp4TnZ8bLc6YwH \
  -H "Authorization: Bearer sk_live_xxx"

export async function POST(request: Request) {
  const callbackData = await request.json();

  if (callbackData.status === "completed") {
    const videoUrl = callbackData.data.results[0];
    // Save the video URL or update your own task record here.
  }

  if (callbackData.status === "failed") {
    const errorMessage = callbackData.data.failed_reason;
    // Mark your own task record as failed here.
  }

  return new Response(null, { status: 200 });
}

Create a video task with POST /v1/videos/generations. The request body has a top-level model, an optional callback_url, and an input object containing prompt and generation settings.

curl https://api.seedance2.ai/v1/videos/generations \
  -H "Authorization: Bearer sk_live_xxx" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "seedance-2-0",
    "callback_url": "https://your-domain.com/api/seedance/webhook",
    "input": {
      "prompt": "a cat surfing on a neon wave, cinematic lighting",
      "generation_type": "text-to-video",
      "duration": 5,
      "aspect_ratio": "16:9",
      "resolution": "720p",
      "generate_audio": true,
      "watermark": false,
      "web_search": false,
      "return_last_frame": false,
      "seed": -1
    }
  }'

generation_type controls which media inputs are accepted and how the model interprets them.

curl https://api.seedance2.ai/v1/videos/generations \
  -H "Authorization: Bearer sk_live_xxx" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "seedance-2-0",
    "callback_url": "https://your-domain.com/api/seedance/webhook",
    "input": {
      "prompt": "a cinematic drone shot over a futuristic coastal city at sunrise",
      "generation_type": "text-to-video",
      "duration": 5,
      "aspect_ratio": "16:9",
      "resolution": "720p",
      "generate_audio": true,
      "watermark": false,
      "web_search": false,
      "return_last_frame": false,
      "seed": -1
    }
  }'

curl https://api.seedance2.ai/v1/videos/generations \
  -H "Authorization: Bearer sk_live_xxx" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "seedance-2-0",
    "input": {
      "prompt": "the subject turns toward camera, soft studio motion",
      "generation_type": "image-to-video",
      "image_urls": ["https://your.cdn.com/first-frame.jpg"],
      "duration": 5,
      "resolution": "720p"
    }
  }'

curl https://api.seedance2.ai/v1/videos/generations \
  -H "Authorization: Bearer sk_live_xxx" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "seedance-2-0",
    "input": {
      "prompt": "use the product from image 1 and the camera motion from the reference video",
      "generation_type": "reference-to-video",
      "image_urls": ["https://your.cdn.com/product.jpg"],
      "video_urls": ["https://your.cdn.com/camera-motion.mp4"],
      "audio_urls": [],
      "duration": 8,
      "resolution": "1080p"
    }
  }'

Parameter names, enum values, endpoint paths, and examples are part of the API contract. Descriptions below explain how each field behaves.

This is the success response from POST /v1/videos/generations. It means the task has been accepted and credits have been reserved. Use the returned taskId to poll GET /v1/tasks/:id or to match a completion or failure callback.

{
  "taskId": "3f2aK9mR...",
  "credits": 60
}

Use GET /v1/tasks/:id to retrieve the current task state. Poll no more than once every 10 seconds. For production systems, prefer webhooks.

curl https://api.seedance2.ai/v1/tasks/3f2aK9mR7xQp4TnZ8bLc6YwH \
  -H "Authorization: Bearer sk_live_xxx"

{
  "id": "3f2aK9mR...",
  "status": "completed",
  "created_at": 1781234567,
  "model": "seedance-2-0",
  "billing_status": "charged",
  "credits": 60,
  "failed_reason": null,
  "data": {
    "results": ["https://cdn.seedance2.ai/.../x.mp4"],
    "video_expires_at": "2026-06-13T10:00:00Z",
    "last_frame_url": null,
    "processing_time": 48
  }
}

{
  "id": "3f2aK9mR...",
  "status": "failed",
  "created_at": 1781234567,
  "model": "seedance-2-0",
  "billing_status": "refunded",
  "credits": 60,
  "failed_reason": "provider_failed"
}

When callback_url is present, Seedance calls your endpoint when the task is completed or failed, and sends JSON data describing the final result. If your endpoint returns a non-2xx response or does not respond within 15 seconds, delivery is retried up to 5 times. Retries reuse the same task id, so de-dupe by id. Return a 200 response as soon as you have safely recorded the callback data.

{
  "id": "3f2aK9mR...",
  "status": "completed",
  "created_at": 1781234567,
  "model": "seedance-2-0",
  "data": {
    "results": ["https://cdn.seedance2.ai/.../x.mp4"],
    "video_expires_at": "2026-06-13T10:00:00Z",
    "last_frame_url": null,
    "processing_time": 48
  }
}

{
  "id": "3f2aK9mR...",
  "status": "failed",
  "created_at": 1781234567,
  "model": "seedance-2-0",
  "data": {
    "failed_reason": "provider_failed",
    "credits_refunded": 60
  }
}

export async function POST(request: Request) {
  const callbackData = await request.json();

  if (callbackData.status === "completed") {
    const videoUrl = callbackData.data.results[0];
    // Save the video URL or update your own task record here.
  }

  if (callbackData.status === "failed") {
    const errorMessage = callbackData.data.failed_reason;
    // Mark your own task record as failed here.
  }

  return new Response(null, { status: 200 });
}

POST /v1/videos/generations and GET /v1/tasks/:id return this error shape when the API request itself fails, such as invalid parameters, invalid API key, insufficient credits, rate limiting, or task not found. Some errors include extra fields such as required, available, or retry_after depending on the situation.

{
  "error": {
    "code": "insufficient_credits",
    "message": "Not enough credits for this task.",
    "required": 60,
    "available": 12
  }
}

Rate limits are applied per API key with a sliding window. Generation defaults to 60 requests per minute; status queries are more lenient. HTTP 429 responses include Retry-After.

The API uses reserve on submit, charge on success, and refund on failure. Dashboard usage pages show API credit history, task logs, and time-based usage statistics.