Qwen text to image

For the text-to-image models in the final Ali whitelist: qwen-image-2.0-pro and wan2.7-image-pro.

Public fields: model, prompt, n, size, response_format, seed, watermark, prompt_extend, and negative_prompt.

POST /v1/images/generations

cURL

basic_request

curl --request POST \ --url https://api.powertokens.ai/v1/images/generations \ --header 'Authorization: Bearer <token>' \ --header 'Content-Type: application/json' \ --data ' { "model": "qwen-image-2.0-pro", "prompt": "A cute orange cat sitting on a windowsill, sunlight shining on it, realistic style", "n": 1, "size": "1024*1024", "response_format": "url" } '

200

{
  "created": 123,
  "data": [
    {
      "url": "<string>",
      "b64_json": "<string>",
      "revised_prompt": "<string>"
    }
  ],
  "metadata": {}
}

Authorizations

Authorization · string · header · required

Bearer authentication header of the form Bearer <token>, where <token> is your auth token.

Body

application/json

model · enum<string> · required

Image generation model name.

Available options: qwen-image-2.0-pro, wan2.7-image-pro

prompt · string · required

Image generation prompt describing desired content, style, and composition. Supports Chinese and English, max 800 characters.

n · integer · default: 1

Number of images to generate. qwen-image-2.0 series supports 1-6, other models fixed at 1.

Required range: 1 <= x <= 6

size · string

Output image resolution in format width*height. qwen-image-2.0 series: total pixels between 512512 and 20482048. qwen-image-max/plus series: supports 1664928(16:9), 14721104(4:3), 13281328(1:1), 11041472(3:4), 928*1664(9:16)

Example:

"1024*1024"

response_format · enum<string> · default: url

Response format. url returns download link, b64_json returns Base64 encoded data.

Available options: url, b64_json

seed · integer

Random seed, range [0, 2147483647]. Same seed produces similar results.

watermark · boolean · default: false

Whether to add watermark to the image.

prompt_extend · boolean · default: true

Whether to enable prompt intelligent rewriting.

negative_prompt · string

Negative prompt describing content to avoid. Supports Chinese and English, max 500 characters.

Response

200

application/json

Success, returns image generation result.

created · integer<int64>

Unix timestamp of creation time (seconds).

data · object[]

List of generated images.

Show child attributes

metadata · object

Additional result metadata. Returned fields may vary by capability.