Description

This interface generates videos from input text and images, including text-to-video and image-to-video.

Notes

This interface is in invitation-only testing. Contact your account manager to use it.

Request syntax

POST /?job_type=VideoGeneration&media_jobs HTTP/1.1​
Host: {BucketName}.tos-cn-beijing.volces.com​
Date: Fri, 30 Jul 2021 08:05:36 +0000​
Content-Type: application/json​
Content-Length: <length>​
Authorization: authorization string​
​
{​
  "GenerationConfig": {​
    "Model": {​
      "Name": "doubao-seedance-1.0-pro"​
    },​
    "Prompt": {​
      "Text": "A woman in a green sequined dress stands against a pink backdrop, surrounded by colorful confetti.",​
      "Frames":  [​
          "first-frame.png",​
          "last-frame.png"​
      ]​
    },​
    "Video": {​
      "Format": "mp4",​
      "Resolution": "720p",​
      "Duration": 5,​
      "Ratio": "16:9",​
      "Fps": 24,​
      "Watermark": true,​
      "CameraFixed": true​
    }​
  },​
  "Output": {​
    "Region": "cn-beijing",​
    "Bucket": "testbkt",​
    "Object": "output.mp4"​
  }​
}​

Request parameters and headers

For common request headers, see Common Parameters.

Field name	Location	Type	Required	Example value	Description
job_type	Query	String	Yes	VideoGeneration	Task type; in video generation scenarios, the value is only VideoGeneration.
media_jobs	Query	String	Yes	-	Special identifier representing media task requests.

Request body elements

Field name	Type	Required	Example value	Description
Model	Container	Yes	-	Doubao model information.
Name	String	Yes	doubao-seedance-1-0-pro	Model name, currently only supports doubao-seedance-1-0-pro. Parent node: Model
Version	String	No	250528	Model version, Name and Version are used to generate the model ID, for example doubao-seedance-1-0-pro-250528. tip When Version is empty, the system-recommended version is used by default. Parent node: Model
Prompt	Container	Yes	-	Video generation prompt information.
Text	String	Yes	A woman wearing a green sequin dress stands in front of a pink background, surrounded by colorful confetti	Text prompts for the text-to-video feature supporting both Chinese and English, limited to 500 bytes in length. Parent node: Prompt
Frames	[]String	No	[ "first-frame.png", "last-frame.png"]	The image-to-video feature allows specifying images (up to two) using the following methods: Based on the "first frame" method (specify one image) Based on the "first and last frame" method (specify two images) tip The uploaded images must meet the following criteria: Image formats: jpeg, png, webp, bmp, tiff, gif Aspect ratio (width/height): (0.4, 2.5) Dimension (px): (300, 6000) Size: less than 30MB Parent node: Prompt
Video	Container	Yes	-	Output video information.
Format	String	No	mp4	Output video container format, currently fixed to mp4. Parent node: Video
Resolution		No	480p	Output video resolutions, available values are as follows: 480p 720p (default) 1080p tip The output video pixel dimensions vary with different Resolution and Ratio parameter combinations. For details, please refer to Width and height pixel values of output videos with different resolutions and aspect ratios. Parent node: Video
Ratio		No	4:3	Output video aspect ratio. The available values are as follows: 16:9 (default) 4:3 1:1 3:4 9:16 21:9 tip The output video’s pixel dimensions vary according to different Resolution and Ratio parameter combinations. For details, please refer to Width and height pixel values of output videos with different resolutions and aspect ratios. Parent node: Video
Duration	Integer	No	5	Output video duration (seconds), valid range [3, 12]. Parent node: Video
Fps	Integer	No	24	Video frame rate, currently fixed at 24. Parent node: Video
Watermark	Boolean	No	true	Whether the generated video includes a watermark. Parent node: Video
CameraFixed	Boolean	No	true	Whether the camera fixed. Parent node: Video

Width and height pixel values of output videos with different resolutions and aspect ratios

Video resolution	Video aspect ratio
Video resolution		16:9	4:3	1:1	3:4	9:16	21:9
480p	864×480	736×544	640×640	544×736	480×864	960×416
720p	1248×704	1120×832	960×960	832×1120	704×1248	1504×640
1080p	1920×1088	1664×1248	1440×1440	1248×1664	1088×1920	2176×928

Response headers

The response contains common headers. For more information, see Common Parameters.

Response body elements

Field name	Type	Example value	Description
Code	String	OK	Status code.
Message	String	Success	Request status information.
JobId	String	1852a9645b553232cdf10001fd35952a4b08	Video generation task ID.

Sample request

Sample response

Torch Object Storage