Vidu image generation — Create a task

POST

/services/aigc/image-generation/generation

# Singapore (ap-southeast-1). This model is available only in this region.
# API keys are region-specific. See https://docs.qwencloud.com/developer-guides/getting-started/regions
curl --location 'https://ap-southeast-1.maas.qwencloudapi.com/api/v1/services/aigc/image-generation/generation' \
  --header 'X-DashScope-Async: enable' \
  --header "Authorization: Bearer $DASHSCOPE_API_KEY" \
  --header 'Content-Type: application/json' \
  --data '{
  "model": "vidu/vidu-image_reference2image",
  "input": {
    "messages": [
      {
        "role": "user",
        "content": [
          {"text": "A flower shop with ornate windows, a beautiful wooden door, and flowers on display"}
        ]
      }
    ]
  },
  "parameters": {"size": "1024*1024", "n": 1, "watermark": false}
}'

{
  "output": {
    "task_status": "PENDING",
    "task_id": "0385dc79-5ff8-4d82-bcb6-xxxxxx"
  },
  "request_id": "4909100c-7b5a-9f92-bfe5-xxxxxx"
}

vidu/vidu-image_reference2image generates images from text, edits an existing image, or generates an image from one or more reference images. It renders Chinese and English text precisely and reproduces UI, chart, and design details at the pixel level, which makes it suitable for posters and infographics. Supported regions: Singapore only. Use an API key created in that region and the https://ap-southeast-1.maas.qwencloudapi.com endpoint. See Regions and endpoints.

Prompt: up to 5,000 characters. At least one non-empty text is required.
Reference images: up to 14 images. PNG, JPG, or WEBP; aspect ratio 1:4 to 4:1; total size up to 50 MB.
Output: one PNG image per task. Supported sizes are 1K, 2K, and 4K.

Processing is asynchronous: this operation returns a task_id, and you poll Query the result until the task finishes. Image generation typically takes 10 to 60 seconds.

Authorizations

string

header

required

QwenCloud API key. See Get an API key.

Header Parameters

enum<string>

required

The content type of the request. Must be application/json.

Available options:application/json

enum<string>

required

Enables asynchronous processing. HTTP requests support only asynchronous calls, so this must be enable. If the header is missing, the request fails with current user api does not support synchronous calls.

Available options:enable

Body

application/json

enum<string>

required

The model name.

Available options:vidu/vidu-image_reference2image

Example:vidu/vidu-image_reference2image

object

required

The input object.

Show child attributes

object[]

required

A message list with exactly one object that contains role and content. The server uses the first non-empty text as the prompt and all image entries as reference images.

Show child attributes

string

The role of the message. Set this to user.

Example:user

object[]

The message content. Contains the text prompt and optional reference images (up to 14).

Show child attributes

string

A positive prompt that describes the content, style, and composition of the image. Supports Chinese and English, up to 5,000 characters. At least one non-empty text is required.

Example:A flower shop with ornate windows, a beautiful wooden door, and flowers on display

string

The URL of a reference image. Supports HTTP and HTTPS. Up to 14 reference images. Formats: PNG, JPG, WEBP. Aspect ratio: 1:4 to 4:1. Total file size: up to 50 MB.

Example:https://example.com/reference.png

object

Image generation parameters.

Show child attributes

string

default"1024*1024"

The image size in width*height format. Defaults to 1024*1024. Supported values by tier:

1K: 10241024, 7201440, 1440720, 1024768, 7681024, 19201088, 10881920, 15361024, 10241536, 1920816, 816*1920
2K: 20482048, 10882160, 21601088, 27362048, 20482736, 25601440, 14402560, 30722048, 20483072, 25601104, 1104*2560
4K: 28802880, 14402880, 28801440, 33122480, 24803312, 38402160, 21603840, 35202352, 23523520, 38401648, 1648*3840

Example:2048*2048

enum<integer>

default1

The number of images to generate. Only 1 is supported.

Available options:1

integer

The random seed. Range: 0 to 2147483647. 0 means random. The same seed produces similar results.

Required range:0 <= x <= 2147483647

boolean

defaultfalse

Whether to add a watermark. Default: false.

Response

200-application/json

string

Unique request identifier for tracing and troubleshooting.

object

Show child attributes

string

The task ID. It is valid for queries for 24 hours.

enum<string>

The task status. New tasks start as PENDING.

Available options:PENDING,RUNNING,SUCCEEDED,FAILED,CANCELED,UNKNOWN

string

Error code. Returned only when the request fails.

string

Detailed error message. Returned only when the request fails.