Wan — Edit a video

POST

/services/aigc/video-generation/video-synthesis

curl --location 'https://dashscope-intl.aliyuncs.com/api/v1/services/aigc/video-generation/video-synthesis' \
--header 'X-DashScope-Async: enable' \
--header "Authorization: Bearer $DASHSCOPE_API_KEY" \
--header 'Content-Type: application/json' \
--data '{
  "model": "wan2.1-vace-plus",
  "input": {
    "function": "image_reference",
    "prompt": "In the video, a girl gracefully walks out from a misty, ancient forest. Her steps are light, and the camera captures her every nimble moment. When she stops and looks around at the lush woods, a smile of surprise and joy blossoms on her face. This scene, frozen in a moment of interplay between light and shadow, records her wonderful encounter with nature.",
    "ref_images_url": [
      "http://wanx.alicdn.com/material/20250318/image_reference_2_5_16.png",
      "http://wanx.alicdn.com/material/20250318/image_reference_1_5_16.png"
    ]
  },
  "parameters": {
    "prompt_extend": true,
    "obj_or_bg": ["obj","bg"],
    "size": "1280*720"
  }
}'

{
  "request_id": "<string>",
  "output": {
    "task_id": "<string>",
    "task_status": "PENDING"
  }
}

This model generates videos from text, image, and video inputs.

Authorizations

string

header

required

DashScope API Key. Create one in the Qwen Cloud console.

Header Parameters

enum<string>

required

Must be enable to create an asynchronous task.

Available options:enable

Body

application/json

enum<string>

required

Model name.

Available options:wan2.1-vace-plus

Example:wan2.1-vace-plus

object

required

Input data for video editing. Required fields vary by function.

Show child attributes

enum<string>

required

The video editing capability to use.

Available options:image_reference,video_repainting,video_edit,video_extension,video_outpainting

Example:image_reference

string

required

Text description of the desired video content.

Example:In the video, a girl walks out from the depths of an ancient, misty forest.

string[]

Array of reference image URLs. Used by image_reference (1-3 images) and optionally by video_repainting and video_edit (1 image to replace subject).

Example:

[
  "http://wanx.alicdn.com/material/20250318/image_reference_2_5_16.png",
  "http://wanx.alicdn.com/material/20250318/image_reference_1_5_16.png"
]

Required range:items: 1–3

string

URL of the input video. Required for video_repainting, video_edit, and video_outpainting. Must be MP4 format, no larger than 50 MB, and no longer than 5 seconds.

Example:http://wanx.alicdn.com/material/20250318/video_repainting_1.mp4

string

URL of a mask image for video_edit. White areas are edited; black areas remain unchanged. Specify either this or mask_video_url.

string

URL of a mask video for video_edit. Specify either this or mask_image_url. Using mask_image_url is recommended.

integer

default0

For video_edit with mask_image_url: specifies which video frame the mask corresponds to. Default: first frame (0).

Example:1

string

For video_extension: URL of the first video clip (3 seconds or shorter). The model generates the rest of the video based on this clip.

string

For video_extension: URL of the last video clip (3 seconds or shorter). The model generates the preceding content.

string

For video_extension: URL of the first frame image. The video extends forward from this frame.

string

For video_extension: URL of the last frame image. Generation proceeds backward from this frame.

object

Generation parameters for video editing. Available parameters vary by function.

Show child attributes

boolean

defaulttrue

Enable prompt rewriting. true (default): model optimizes your prompt. false: use your prompt as-is. We recommend disabling this for video_repainting.

string

Output resolution as width*height. Used by image_reference and video_edit.

Example:1280*720

enum<string>[]

For image_reference: identifies each reference image as a subject (obj) or background (bg). Array length must match ref_images_url length.

Example:

[
  "obj",
  "bg"
]

enum<string>

For video_repainting: video feature extraction method. Determines which features from the original video are retained.

Available options:posebodyface,posebody,depth,scribble

Example:depth

number

default1

For video_repainting: controls feature extraction strength. Higher values make the output more similar to the original; lower values allow more creative freedom.

Required range:0 <= x <= 1

enum<string>

default"tracking"

For video_edit: specifies the behavior of the editing area. tracking (default): the editing area automatically follows the target's motion. fixed: the editing area stays in a fixed position.

Available options:tracking,fixed

number

default0.05

For video_edit with mask_type: tracking: the ratio by which the mask area expands outward. Lower values fit the target more closely; higher values expand the mask area.

Required range:0 <= x <= 1

number

default1

For video_outpainting: upward expansion ratio. Setting to 1.5 expands the top to 1.5 times original height.

Required range:1 <= x <= 2

number

default1

For video_outpainting: downward expansion ratio.

Required range:1 <= x <= 2

number

default1

For video_outpainting: leftward expansion ratio.

Required range:1 <= x <= 2

number

default1

For video_outpainting: rightward expansion ratio.

Required range:1 <= x <= 2

Response

200-application/json

string

Unique request identifier.

object

Show child attributes

string

Task identifier. Use this with GET /tasks/{task_id} to poll for results.

enum<string>

Initial task status, typically PENDING.

Available options:PENDING