Image To Video Generation

Image-to-Video API generates desired videos via images and keywords, leveraging the latest models to produce tailored visual outputs effectively.

API documentation

Interface for removing background from the submitted video：

Request URL: https://www.cutout.pro/api/v2/imageToVideo/generate
Request method: POST
Return type: json data
Input parameters:

(1)Request parameters (Header)

Parameter	Parameter Type	Description
APIKEY	string	Your API Key

(2) Payload

Parameter	Parameter Type	Description
imageUrl	string	Image URL, width and height restrictions: minimum 300 pixels, maximum no more than 6000 pixels, and image size cannot exceed 10MB.
prompt	string	Keywords (optional). Filling in keywords can make the video closer to the effect you want.

(3) Response data

 {
    "code":0, // 0 indicates success, and the other values are the corresponding error codes
    "data": 700842286474565, 
    "msg":null, // if code is not 0, the error msg of the error
    "time": 1615368038661, // the unix timestamp of server
    "requestId": "6880539bcc69f3e9bd67a5a6fb35d0fc"
 }

Sample Code

bash

curl -X POST "https://www.cutout.pro/api/v2/imageToVideo/generate" \
     -H "APIKEY: The API key of the account" \
     -H "Content-Type: application/json" \
     -d '{
          "imageUrl": "image url",
          "prompt": "prompt"
      }'

python

  import requests
  import json

  url = "https://www.cutout.pro/api/v2/imageToVideo/generate"
  headers = {
      "APIKEY": "The API key of the account",
      "Content-Type": "application/json"
  }
  data = {
      "imageUrl": "image url",
      "prompt": "prompt"
  }

  response = requests.post(url, headers=headers, json=data)
  print(response.text)

php

  <?php
  $url = 'https://www.cutout.pro/api/v2/imageToVideo/generate';
  $data = array(
      'imageUrl' => 'image url',
      'prompt' => 'prompt'
  );

  $options = array(
      'http' => array(
          'method' => 'POST',
          'header' => "APIKEY: The API key of the account\r\n" .
                      "Content-Type: application/json\r\n",
          'content' => json_encode($data)
      )
  );

  $context = stream_context_create($options);
  $result = file_get_contents($url, false, $context);
  echo $result;
  ?>

java

  import org.springframework.http.HttpEntity;
  import org.springframework.http.HttpHeaders;
  import org.springframework.http.MediaType;
  import org.springframework.web.client.RestTemplate;

  public class RequestExample {
      public static void main(String[] args) {
          RestTemplate restTemplate = new RestTemplate();
          String url = "https://www.cutout.pro/api/v2/imageToVideo/generate";
  
          // setting header
          HttpHeaders headers = new HttpHeaders();
          headers.set("APIKEY", "The API key of the account");
          headers.setContentType(MediaType.APPLICATION_JSON);
  
          // request body json
          String jsonBody = "{\"imageUrl\":\"image url\",\"prompt\":\"prompt\"}";
          HttpEntity<String> request = new HttpEntity<>(jsonBody, headers);
  
          // send post request
          String response = restTemplate.postForObject(url, request, String.class);
          System.out.println(response);
      }
  }

nodejs

  const https = require('https');

  const postData = JSON.stringify({
    "imageUrl": "https://example.com/your-image.jpg",
    "prompt": "prompt" 
  });

  const options = {
    hostname: 'www.cutout.pro',
    path: '/api/v2/imageToVideo/generate',
    method: 'POST',
    headers: {
      'APIKEY': 'your_api_key_here',
      'Content-Type': 'application/json',
      'Content-Length': Buffer.byteLength(postData)
    }
  };

  const req = https.request(options, (res) => {
    let responseData = '';
  
    res.on('data', (chunk) => {
      responseData += chunk;
    });
   
    res.on('end', () => {
      console.log('request success：');
      console.log(JSON.parse(responseData));
    });
  });

  req.on('error', (e) => {
    console.error('request error：', e.message);
  });
  
  req.write(postData);
  req.end();

Query generation results：

Request URL: https://www.cutout.pro/api/v2/imageToVideo/getResult
Request method: GET
Input parameters:

(1) Request parameters (Header)

Parameter	Parameter Type	Description
APIKEY	string	Your API key

(2) Query String

Parameter	Parameter Type	Description
taskId	long	The task ID returned by the submission endpoint.

(3) Response data

{
    "code": 0,
    "data": {
        "id": 700842286474565,
        "userId": 528874,
        "imageUrl": "https://example.com/your-image.jpg",
        "originalPrompt": "Request keywords",
        "resultUrl": "The resulting video URL",
        "cover": "The video cover URL of the generated result",
        "width": 1024,
        "height": 1024,
        "status": 1, //0 = process | 1 = success | 2 = failed
        "percentage": 100,
        "waitNumber": 0,
        "createdAt": 1753240476000
    },
    "msg": "",
    "time": 1753241547143,
    "requestId": "688057cb0234b17e58a64906f583ae12"
}

Return result parameter:

id	The task id returned when generating.
userId	The id used by the current APIKEY.
imageUrl	The image URL requested when generating.
originalPrompt	Keywords requested when generating.
resultUrl	Generate the resulting URL for the video.
cover	Generate video cover.
width	Generate the width of the video, calculated based on the video cover
height	The height of the generated video is calculated based on the video cover
status	The status of the generated video, 0 = generating, 1 = generating successfully, 2 = generating failed
percentage	The progress of processing the video. When the progress is 100, the video generation is complete.
waitNumber	The number of requests currently waiting in queue
createdAt	Time when the video was created

(4) Error Reponse

{
    "code": 9011,
    "data": null,
    "msg": "Image contains inappropriate content, cannot be processed.",
    "time": 1753241547143,
    "requestId": "688057cb0234b17e58a64906f583ae12"
}

Error Code:

Code	Msg
9011	Image contains inappropriate content, cannot be processed.
1001	The server is abnormal. Please contact us.
4002	No results were found for the current task ID. Please check whether `taskId` is correct.

Sample Code

bash

curl -H 'APIKEY: INSERT_YOUR_API_KEY_HERE' \
  'https://www.cutout.pro/api/v2/imageToVideo/getResult?taskId=1111'

python

  import requests

  api_key = "your_api_key_here"
  task_id = "your_task_id_here"
  url = f"https://www.cutout.pro/api/v2/imageToVideo/getResult?taskId={task_id}"
  
  headers = {
      "APIKEY": api_key
  }

  response = requests.get(url, headers=headers)
  print(response.text)

php

  <?php
  $apiKey = "your_api_key_here";
  $taskId = "your_task_id_here";
  $url = "https://www.cutout.pro/api/v2/imageToVideo/getResult?taskId=$taskId";
  
  $headers = array(
      'APIKEY: ' . $apiKey
  );
  
  $ch = curl_init();
  curl_setopt($ch, CURLOPT_URL, $url);
  curl_setopt($ch, CURLOPT_HTTPHEADER, $headers);
  curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
  
  $response = curl_exec($ch);
  curl_close($ch);
  
  echo $response;
  ?>

java

  import org.springframework.http.HttpEntity;
  import org.springframework.http.HttpHeaders;
  import org.springframework.http.HttpMethod;
  import org.springframework.http.ResponseEntity;
  import org.springframework.web.client.RestTemplate;

  public class Main {
      public static void main(String[] args) {
          RestTemplate restTemplate = new RestTemplate();
          String apiKey = "your_api_key_here";
          String taskId = "your_task_id_here";
          String url = "https://www.cutout.pro/api/v2/imageToVideo/getResult?taskId=" + taskId;
  
          HttpHeaders headers = new HttpHeaders();
          headers.set("APIKEY", apiKey);
  
          HttpEntity<String> entity = new HttpEntity<>(headers);
  
          ResponseEntity<String> response = restTemplate.exchange(
                  url,
                  HttpMethod.GET,
                  entity,
                  String.class
          );
  
          System.out.println(response.getBody());
      }
  }

nodejs

  const https = require('https');

  const apiKey = "your_api_key_here";
  const taskId = "your_task_id_here";
  const options = {
      hostname: 'www.cutout.pro',
      path: `/api/v2/imageToVideo/getResult?taskId=${taskId}`,
      method: 'GET',
      headers: {
          'APIKEY': apiKey
      }
  };

  const req = https.request(options, (res) => {
      let data = '';
  
      res.on('data', (chunk) => {
          data += chunk;
      });
  
      res.on('end', () => {
          console.log(data);
      });
  
  }).on('error', (error) => {
      console.error(error);
  });
  
  req.end();

The cost incurred by the request

Each request will consume 5 video credits.

Seedance 2.0 and Seedance 2.5 Video Generation

The Seedance API supports text-to-video, first-frame, first-and-last-frame, multimodal reference, video editing, and reference-video continuation requests.

Calling sequence

Seedance generation is an asynchronous workflow. Call the APIs in the following order:

Call /seedance2/uploadByUrl once for every image, video, or audio URL used by the task.
Keep the mediaType, assetId, and url returned for each asset.
Put the returned asset objects into the appropriate fields and call /seedance/generate.
Save the task ID returned by the generation API.
Poll /seedance/getResult with the task ID until status becomes 2 or 3.

Important

All three APIs must be called with the same APIKEY. Assets and generated tasks belong to the API user that created them. Seedance API access is currently available only to authorized API users.

1. Import a Seedance asset by URL

This API downloads a publicly accessible image, video, or audio URL, copies the file to private storage, and creates a Seedance asset.

Request URL: https://www.cutout.pro/api/v2/imageToVideo/seedance2/uploadByUrl
Request method: POST
Content type: application/x-www-form-urlencoded
Return type: JSON

Request parameters

Parameter	Type	Required	Description
APIKEY	string	Yes	Your API key. The API user must be authorized to use Seedance asset upload.

Form parameters

Parameter	Type	Required	Description
url	string	Yes	A public HTTP or HTTPS URL that the server can download directly.
mediaType	string	Yes	Asset type: `image`, `video`, or `audio`. It must match the actual file content.

cURL examples

Import an image:

bash

curl --request POST \
  --url 'https://www.cutout.pro/api/v2/imageToVideo/seedance2/uploadByUrl' \
  --header 'APIKEY: INSERT_YOUR_API_KEY_HERE' \
  --header 'Content-Type: application/x-www-form-urlencoded' \
  --data-urlencode 'url=https://example.com/reference-image.jpg' \
  --data-urlencode 'mediaType=image'

Import a video:

bash

curl --request POST \
  --url 'https://www.cutout.pro/api/v2/imageToVideo/seedance2/uploadByUrl' \
  --header 'APIKEY: INSERT_YOUR_API_KEY_HERE' \
  --header 'Content-Type: application/x-www-form-urlencoded' \
  --data-urlencode 'url=https://example.com/reference-video.mp4' \
  --data-urlencode 'mediaType=video'

Import an audio file:

bash

curl --request POST \
  --url 'https://www.cutout.pro/api/v2/imageToVideo/seedance2/uploadByUrl' \
  --header 'APIKEY: INSERT_YOUR_API_KEY_HERE' \
  --header 'Content-Type: application/x-www-form-urlencoded' \
  --data-urlencode 'url=https://example.com/reference-audio.mp3' \
  --data-urlencode 'mediaType=audio'

Successful response example

json

{
  "code": 0,
  "data": {
    "taskFlag": null,
    "mediaType": "image",
    "url": "https://example-storage.com/upload/reference-image.jpg?X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Signature=SIGNED_VALUE",
    "firstFrameUrl": null,
    "assetId": "asset-20260831123000-abcd1"
  },
  "msg": "",
  "time": 1788141000000,
  "requestId": "example-request-id"
}

Response fields

Field	Type	Description
taskFlag	string or null	Not used by URL import and normally returned as `null`.
mediaType	string	The verified asset type: `image`, `video`, or `audio`.
url	string	The private signed URL created by the service. Pass this value to the generation API without modifying it.
firstFrameUrl	string or null	URL import does not extract a video first frame, so this field is normally `null`. It is not required by the generation API.
assetId	string	The Seedance asset ID created for this file.

Asset requirements

Type	Supported formats	Maximum size	Additional requirements
Image	JPG, JPEG, PNG, WEBP	20 MB	The URL content must be a valid image.
Video	MP4, MOV	50 MB	Duration: 2–15 seconds; width and height: 300–6000 px; total pixels: 409600–927408; aspect ratio: 0.4–2.5; frame rate: 24–60 fps.
Audio	MP3, WAV	15 MB	Duration: 2–15 seconds.

Asset object used by the next API

For the following examples, replace the placeholder values with the complete values returned by this API:

json

{
  "mediaType": "image",
  "assetId": "asset-20260831123000-abcd1",
  "url": "https://example-storage.com/upload/reference-image.jpg?X-Amz-Signature=SIGNED_VALUE",
  "firstFrameUrl": null
}

2. Submit a Seedance generation task

Request URL: https://www.cutout.pro/api/v2/imageToVideo/seedance/generate
Request method: POST
Content type: application/json
Return type: JSON

Request parameters

Header

Parameter	Type	Required	Description
APIKEY	string	Yes	The same API key used to import the assets.

JSON payload

Parameter	Type	Required	Description
prompt	string	Yes	The generation or editing instruction. In reference mode, `@image1`, `@video1`, and `@audio1` refer to assets by their array order.
model	string	Yes	`seedance_2_pro`, `seedance_2_pro_fast`, `seedance_2_0_mini`, or `seedance_2_5`.
resolution	string	Yes	Output resolution. See the model table below.
ratio	string	Yes	Output aspect ratio. Use `adaptive` when the model should infer the ratio from the input.
duration	integer	Yes	Requested duration in seconds. For Seedance 2.5 editing, use `-1`.
generateAudio	boolean	No	Whether the model should generate audio. Defaults to `false` when omitted.
scene	string	No	Business scene identifier. The recommended value is `ai-video-seedance2`.
firstFrameAsset	object	No	The image asset used as the first frame.
lastFrameAsset	object	No	The image asset used as the last frame. It cannot be supplied without `firstFrameAsset`.
referenceImageAssets	array	No	Reference image asset objects.
videoAssets	array	No	Reference or source video asset objects.
audioAssets	array	No	Reference audio asset objects.

Each asset object accepts these fields:

Field	Type	Description
mediaType	string	`image`, `video`, or `audio`.
assetId	string	The `assetId` returned by `/seedance2/uploadByUrl`.
url	string	The complete signed `url` returned by `/seedance2/uploadByUrl`.
firstFrameUrl	string or null	Optional. It may be left as `null` for assets imported by URL.

Supported models

Model	Supported resolution	Duration
seedance_2_pro	480p, 720p, 1080p	A positive duration supported by the model.
seedance_2_pro_fast	480p, 720p	A positive duration supported by the model.
seedance_2_0_mini	480p, 720p	A positive duration supported by the model.
seedance_2_5	480p, 720p, 1080p	4–30 seconds; use `-1` only for video editing.

Seedance pricing

Seedance is billed per generated second. The price depends on the selected model and resolution.

The service uses image credits first. If the account does not have enough image credits, it uses video credits at the conversion rate of 2 image credits = 1 video credit.

Model	Resolution	Image credits per second	Video credits per second
seedance_2_pro	480p	3	1.5
seedance_2_pro	720p	5	2.5
seedance_2_pro	1080p	10	5
seedance_2_pro_fast	480p	2	1
seedance_2_pro_fast	720p	4	2
seedance_2_0_mini	480p	2	1
seedance_2_0_mini	720p	4	2
seedance_2_5	480p	10	5
seedance_2_5	720p	16	8
seedance_2_5	1080p	60	30

For normal generation, calculate the estimated cost as follows:

text

Image-credit cost = image credits per second × requested duration
Video-credit cost = video credits per second × requested duration

Examples:

seedance_2_pro, 720p, 5 seconds: 5 × 5 = 25 image credits, or 2.5 × 5 = 12.5 video credits.
seedance_2_5, 720p, 10 seconds: 16 × 10 = 160 image credits, or 8 × 10 = 80 video credits.

Seedance 2.5 editing billing

When seedance_2_5 video editing uses duration: -1, the service initially reserves the cost of 30 seconds. After successful completion, it uses the actual output duration reported by the model and refunds the unused portion.

For example, a 720p editing request initially reserves 16 × 30 = 480 image credits, or 8 × 30 = 240 video credits. If the actual output is 8 seconds, the final cost is 16 × 8 = 128 image credits, or 8 × 8 = 64 video credits, and the unused amount is returned.

Billing notes

Reference images, reference videos, reference audio, first/last frames, and generateAudio do not independently change the price. Billing is determined by model, resolution, and billable duration.
A failed generation task returns the credits charged for that task according to the service's failure-refund processing.
Billing and refund details are intentionally not included in /seedance/getResult; that endpoint only returns task progress and output information.

Generation mode examples

The examples below use placeholder asset values. Import every source file first and replace both assetId and url with the corresponding upload response.

A. Multimodal reference mode

Reference images, videos, and audio can be combined in one request. Array numbering starts at 1, so the first entries correspond to @image1, @video1, and @audio1 in the prompt.

bash

curl --request POST \
  --url 'https://www.cutout.pro/api/v2/imageToVideo/seedance/generate' \
  --header 'APIKEY: INSERT_YOUR_API_KEY_HERE' \
  --header 'Content-Type: application/json' \
  --data-raw '{
    "scene": "ai-video-seedance2",
    "model": "seedance_2_pro",
    "prompt": "Let the person in @image1 enter the scene in @video1 and move to the rhythm of @audio1.",
    "referenceImageAssets": [
      {
        "mediaType": "image",
        "assetId": "asset-image-001",
        "url": "https://example-storage.com/reference-image.jpg?X-Amz-Signature=SIGNED_VALUE",
        "firstFrameUrl": null
      }
    ],
    "videoAssets": [
      {
        "mediaType": "video",
        "assetId": "asset-video-001",
        "url": "https://example-storage.com/reference-video.mp4?X-Amz-Signature=SIGNED_VALUE",
        "firstFrameUrl": null
      }
    ],
    "audioAssets": [
      {
        "mediaType": "audio",
        "assetId": "asset-audio-001",
        "url": "https://example-storage.com/reference-audio.mp3?X-Amz-Signature=SIGNED_VALUE",
        "firstFrameUrl": null
      }
    ],
    "firstFrameAsset": null,
    "lastFrameAsset": null,
    "resolution": "720p",
    "ratio": "adaptive",
    "duration": 5,
    "generateAudio": true
  }'

If a media type is not needed, omit its array or send an empty array. Do not put an image asset into videoAssets or an audio asset into referenceImageAssets.

B. First-and-last-frame mode

Use two image assets to define the beginning and end of the generated video.

bash

curl --request POST \
  --url 'https://www.cutout.pro/api/v2/imageToVideo/seedance/generate' \
  --header 'APIKEY: INSERT_YOUR_API_KEY_HERE' \
  --header 'Content-Type: application/json' \
  --data-raw '{
    "scene": "ai-video-seedance2",
    "model": "seedance_2_5",
    "prompt": "Create a natural transition from the first frame to the last frame while preserving the character identity and visual style.",
    "firstFrameAsset": {
      "mediaType": "image",
      "assetId": "asset-first-frame-001",
      "url": "https://example-storage.com/first-frame.jpg?X-Amz-Signature=SIGNED_VALUE",
      "firstFrameUrl": null
    },
    "lastFrameAsset": {
      "mediaType": "image",
      "assetId": "asset-last-frame-001",
      "url": "https://example-storage.com/last-frame.jpg?X-Amz-Signature=SIGNED_VALUE",
      "firstFrameUrl": null
    },
    "referenceImageAssets": [],
    "videoAssets": [],
    "audioAssets": [],
    "resolution": "720p",
    "ratio": "adaptive",
    "duration": 5,
    "generateAudio": true
  }'

For first-frame-only generation, keep firstFrameAsset and set lastFrameAsset to null. Supplying a last frame without a first frame is rejected.

C. Video editing mode

Video editing is available with seedance_2_5. Set duration to -1 and provide at least one source video in videoAssets. The completed response reports the actual output duration.

bash

curl --request POST \
  --url 'https://www.cutout.pro/api/v2/imageToVideo/seedance/generate' \
  --header 'APIKEY: INSERT_YOUR_API_KEY_HERE' \
  --header 'Content-Type: application/json' \
  --data-raw '{
    "scene": "ai-video-seedance2",
    "model": "seedance_2_5",
    "prompt": "Replace the sky with a warm sunset while preserving the original subject, camera movement, and composition.",
    "videoAssets": [
      {
        "mediaType": "video",
        "assetId": "asset-edit-video-001",
        "url": "https://example-storage.com/source-video.mp4?X-Amz-Signature=SIGNED_VALUE",
        "firstFrameUrl": null
      }
    ],
    "referenceImageAssets": [],
    "audioAssets": [],
    "firstFrameAsset": null,
    "lastFrameAsset": null,
    "resolution": "720p",
    "ratio": "adaptive",
    "duration": -1,
    "generateAudio": true
  }'

D. Reference-video continuation mode

To request a continuation, provide a reference video, use a positive output duration, and explicitly describe the continuation in the prompt.

bash

curl --request POST \
  --url 'https://www.cutout.pro/api/v2/imageToVideo/seedance/generate' \
  --header 'APIKEY: INSERT_YOUR_API_KEY_HERE' \
  --header 'Content-Type: application/json' \
  --data-raw '{
    "scene": "ai-video-seedance2",
    "model": "seedance_2_5",
    "prompt": "Continue naturally from the final frame of the source video. The character keeps running forward with consistent appearance, lighting, scene style, and camera motion.",
    "videoAssets": [
      {
        "mediaType": "video",
        "assetId": "asset-continuation-video-001",
        "url": "https://example-storage.com/source-video.mp4?X-Amz-Signature=SIGNED_VALUE",
        "firstFrameUrl": null
      }
    ],
    "referenceImageAssets": [],
    "audioAssets": [],
    "firstFrameAsset": null,
    "lastFrameAsset": null,
    "resolution": "720p",
    "ratio": "adaptive",
    "duration": 10,
    "generateAudio": true
  }'

Continuation behavior

The current request contract does not expose a separate extend flag. A positive-duration request containing only a reference video is submitted as reference-video generation. Clearly describe the continuation in the prompt. Frame-perfect continuation depends on the selected model's current capability and is not guaranteed by a separate API mode switch.

Successful generation response

json

{
  "code": 0,
  "data": 701234567890123,
  "msg": "",
  "time": 1788141300000,
  "requestId": "example-request-id"
}

The numeric value in data is the task ID. Generation is asynchronous; this response means that the task was accepted, not that the video has finished.

3. Query the Seedance task result

Request URL: https://www.cutout.pro/api/v2/imageToVideo/seedance/getResult
Request method: GET
Return type: JSON

Request parameters

Header

Parameter	Type	Required	Description
APIKEY	string	Yes	The same API key that submitted the generation task.

Query string

Parameter	Type	Required	Description
taskId	long	Yes	The task ID returned in `data` by `/seedance/generate`.

cURL example

bash

curl --request GET \
  --url 'https://www.cutout.pro/api/v2/imageToVideo/seedance/getResult?taskId=701234567890123' \
  --header 'APIKEY: INSERT_YOUR_API_KEY_HERE'

Processing response example

json

{
  "code": 0,
  "data": {
    "id": 701234567890123,
    "status": 1,
    "percentage": 46,
    "resultUrl": null,
    "duration": 5,
    "width": null,
    "height": null,
    "failReason": null,
    "runningTime": null,
    "totalTime": null,
    "createdAt": 1788141300000
  },
  "msg": "",
  "time": 1788141330000,
  "requestId": "example-request-id"
}

Successful response example

json

{
  "code": 0,
  "data": {
    "id": 701234567890123,
    "status": 2,
    "percentage": 100,
    "resultUrl": "https://example-storage.com/result/video.mov?X-Amz-Signature=SIGNED_VALUE",
    "duration": 5,
    "width": 1280,
    "height": 720,
    "failReason": null,
    "runningTime": 82340,
    "totalTime": 90125,
    "createdAt": 1788141300000
  },
  "msg": "",
  "time": 1788141391000,
  "requestId": "example-request-id"
}

Failed response example

json

{
  "code": 0,
  "data": {
    "id": 701234567890123,
    "status": 3,
    "percentage": 38,
    "resultUrl": null,
    "duration": 5,
    "width": null,
    "height": null,
    "failReason": "generate video failed",
    "runningTime": 30120,
    "totalTime": 35640,
    "createdAt": 1788141300000
  },
  "msg": "",
  "time": 1788141336000,
  "requestId": "example-request-id"
}

Result fields

Field	Type	Description
id	long	Task ID.
status	integer	`0`: pending; `1`: processing; `2`: succeeded; `3`: failed. Stop polling when the value is `2` or `3`.
percentage	integer	Generation progress from 0 to 100. Use `status`, rather than progress alone, to determine the final state.
resultUrl	string or null	Signed output video URL. Available when the task succeeds.
duration	integer	Video duration in seconds. For completed Seedance 2.5 editing tasks, this is updated to the actual output duration when provided by the model.
width	integer or null	Output video width, available after successful generation.
height	integer or null	Output video height, available after successful generation.
failReason	string or null	User-facing failure reason when `status` is `3`.
runningTime	long or null	Model processing time in milliseconds.
totalTime	long or null	Total backend processing time in milliseconds.
createdAt	long	Task creation time as a Unix timestamp in milliseconds.

Important notes

Import every source URL with /seedance2/uploadByUrl before generating. Do not use an asset created by a different API user.
Preserve the complete signed asset URL, including its query string. Do not decode, shorten, or remove signature parameters.
Pass both assetId and url from the upload response. This keeps the request compatible with the supported Seedance 2.0 and 2.5 models.
Use firstFrameAsset, lastFrameAsset, referenceImageAssets, videoAssets, and audioAssets according to the actual media type. Do not put the same upload response into an unrelated field.
If an asset array is present and non-empty, it takes priority over its legacy raw-URL field. Avoid mixing referenceImageAssets with referenceImages, videoAssets with referenceVideos, or audioAssets with referenceAudios in the same request.
lastFrameAsset requires firstFrameAsset.
Seedance 2.5 edit mode requires duration: -1 and at least one video asset. Other generation modes must use a positive duration.
generateAudio is false when omitted. Set it explicitly when audio generation is required.
Poll every few seconds rather than sending continuous requests. Stop polling on status: 2 or status: 3.
The result response intentionally does not expose credits, refund details, prompts, reference assets, or internal generation parameters.
resultUrl is a signed URL. Download or copy the result promptly, and query the result API again if a refreshed URL is needed.

Image To Video Generation ​

API documentation ​

Interface for removing background from the submitted video： ​

Sample Code ​

Query generation results： ​

Return result parameter: ​

Error Code: ​

Sample Code ​

The cost incurred by the request ​

Seedance 2.0 and Seedance 2.5 Video Generation ​

Calling sequence ​

1. Import a Seedance asset by URL ​

Request parameters ​

Header ​

Form parameters ​

cURL examples ​

Successful response example ​

Response fields ​

Asset requirements ​

2. Submit a Seedance generation task ​

Request parameters ​

Header ​

JSON payload ​

Supported models ​

Seedance pricing ​

Seedance 2.5 editing billing ​

Generation mode examples ​

A. Multimodal reference mode ​

B. First-and-last-frame mode ​

C. Video editing mode ​

D. Reference-video continuation mode ​

Successful generation response ​

3. Query the Seedance task result ​

Request parameters ​

Header ​

Query string ​

cURL example ​

Processing response example ​

Successful response example ​

Failed response example ​

Result fields ​

Important notes ​

Image To Video Generation

API documentation

Interface for removing background from the submitted video：

Sample Code

Query generation results：

Return result parameter:

Error Code:

Sample Code

The cost incurred by the request

Seedance 2.0 and Seedance 2.5 Video Generation

Calling sequence

1. Import a Seedance asset by URL

Request parameters

Header

Form parameters

cURL examples

Successful response example

Response fields

Asset requirements

2. Submit a Seedance generation task

Request parameters

Header

JSON payload

Supported models

Seedance pricing

Seedance 2.5 editing billing

Generation mode examples

A. Multimodal reference mode

B. First-and-last-frame mode

C. Video editing mode

D. Reference-video continuation mode

Successful generation response

3. Query the Seedance task result

Request parameters

Header

Query string

cURL example

Processing response example

Successful response example

Failed response example

Result fields

Important notes