多元探索 API多元探索 API
Qwen Image

Qwen Image Image Generation API

Use `POST /v1/images/generations` to call `qwen-image-3.0` and `qwen-image-3.0-pro` for image generation, supporting both the native multimodal structure and the OpenAI simplified structure.

POSThttps://zx1.deepwl.net/v1/images/generations
Request
curl --request POST 'https://zx1.deepwl.net/v1/images/generations' \
  --header 'Authorization: Bearer YOUR_API_KEY' \
  --header 'Content-Type: application/json' \
  --data '{
    "model": "qwen-image-3.0",
    "input": {
      "messages": [
        {
          "role": "user",
          "content": [
            {
              "text": "A warm film-style street portrait in the afternoon, a person holding a bouquet of roses, backlit, shallow depth of field"
            }
          ]
        }
      ]
    },
    "parameters": {
      "prompt_extend": true,
      "size": "1080*1080",
      "n": 1,
      "watermark": false,
      "seed": 123456
    }
  }'
Response
{
  "created": 1788188983,
  "data": [
    {
      "url": "https://example.com/generated.png"
    }
  ],
  "metadata": {
    "output": {
      "choices": [
        {
          "finish_reason": "stop",
          "message": {
            "role": "assistant",
            "content": [
              {
                "type": "image",
                "image": "https://example.com/generated.png"
              }
            ]
          }
        }
      ]
    },
    "request_id": "example-request-id",
    "usage": {
      "input_image_count": 0,
      "input_image_type": "qima_input_1k",
      "output_image_count": 1,
      "output_image_type": "qima_output_1k",
      "output_width": 1080,
      "output_height": 1080
    }
  }
}

qwen-image-3.0 and qwen-image-3.0-pro use the unified text-to-image entry point (POST /v1/images/generations).

  • When prompt_extend or seed is needed, passing the native multimodal structure directly is recommended.
  • When only basic parameters are needed, the OpenAI simplified structure can also be used.

Request Headers

Authorizationstring必填

Request authentication. Uses a Bearer token, e.g. Bearer YOUR_API_KEY.

Content-Typestring必填

Request content type, must be application/json.

Request Body

modelstring必填

Must be qwen-image-3.0 or qwen-image-3.0-pro.

promptstring

Generation prompt used in the simplified format; optional when input.messages is provided.

inputobject

The native multimodal input structure containing messages. When provided, the top-level prompt can be empty; at least one valid text must be provided between input and the top-level prompt.

messagesarray必填

Message list; pass only one message.

rolestring必填

Message role, fixed to user.

contentarray必填

Message content list. Each content element is either text or image. Text-to-image passes only one text; image-to-image/editing passes 1–3 image items and one text.

textstring

Text content, i.e. the generation prompt or edit instruction. Text-to-image must pass exactly one.

imagestring

Reference image address, supporting HTTP/HTTPS URLs or Data URLs with a MIME prefix. Image-to-image/editing passes 1–3; supported formats: JPG, JPEG, PNG, BMP, TIFF, WEBP, GIF; officially recommended width and height between 384 and 2048 pixels; a single image must not exceed 10 MB.

parametersobject

Qwen image generation parameters. When explicitly provided, the adapter uses these parameters directly and no longer copies size, n, or watermark from the top level.

sizestring

Output size in the native format WIDTH*HEIGHT (e.g. 1080*1080). The simplified format uses the top-level size in WIDTHxHEIGHT.

ninteger

Number of output images; the current upstream range is 1–6. When omitted or 0, it is treated as 1.

watermarkboolean

Whether to add a watermark.

prompt_extendboolean

Whether to enable prompt extension.

seedinteger

Random seed.

ninteger

Number of images in the simplified format. When omitted or 0, it is treated as 1.

sizestring

Output size in the simplified format, using WIDTHxHEIGHT (e.g. 1080x1080).

watermarkboolean

Watermark switch in the simplified format.

OpenAI simplified structure

When only basic parameters are needed, prompt, size, n, and watermark can be passed at the top level:

{
  "model": "qwen-image-3.0",
  "prompt": "A warm film-style street portrait in the afternoon",
  "size": "1080x1080",
  "n": 1,
  "watermark": false
}

The adapter performs the following conversions:

  • Converts prompt to input.messages[0].content[0].text.
  • Converts the x in size to the native format *.
  • When n is omitted or 0, sets n to 1.

Request Example

Response Example

Response Fields

createdinteger

Unix timestamp of when AIX processed the request.

dataarray<object>

OpenAI-style array of image results.

urlstring

Image URL returned by the upstream.

revised_promptstring

Appears when the upstream returns text content.

metadataobject

The original upstream response.

outputobject

The original upstream output structure, where choices[].message.content[].image is the actual returned image address. When data is empty, combine this field with request_id to verify whether the upstream returned an image.

request_idstring

Upstream request ID.

usageobject

Input/output image counts and resolution information.

errorobject

Error structure returned when the request fails (e.g. HTTP 400).

messagestring

Error description.

typestring

Error type, currently fixed to ali_error.

paramstring

The failing parameter name, or an empty string if none.

codestring

Error code.