Qwen-Image-Max (text-to-image)

POST

/aliyun/api/v1/services/aigc/multimodal-generation/generation

Qwen-Image is a versatile text-to-image generation model that supports multiple artistic styles, with particular strengths in complex text rendering. The model enables multi-line layouts, paragraph-level text generation, and fine-grained detail depiction, making it capable of realizing sophisticated text-image hybrid layout designs.

📌 Parameter Specifications

Parameter Name	Data Type	Value Rules & Constraints	Function Description & Details	Default Value
negative_prompt	string	Supports both Chinese and English Max 500 characters; excess will be automatically truncated	Negative prompt: restricts content that should not appear in the generated image	None (recommended to be specified)
prompt_extend	boolean	Only accepts `true` / `false`	Positive prompt intelligent refinement switch: optimizes and polishes only the positive prompt without modifying the negative prompt	✅ `true` (enabled)
watermark	boolean	Only accepts `true` / `false`	Watermark toggle: adds a "Qwen-Image" watermark at the bottom-right corner of the image	❌ `false` (disabled)
size	string	Fixed format: widthheight Available options👇 ▫️ 1664928 (16:9) ▫️ 14721104 (4:3) ▫️ 13281328 (1:1) ▫️ 11041472 (3:4) ▫️ 9281664 (9:16)	Output image resolution: must strictly match the specified values	1328*1328 (1:1)
n	integer	Fixed to `1`; setting other values will trigger an error	Number of images to generate: the current API only supports single-image generation	1
seed	integer	Value range: [0, 2147483647] A random value will be assigned if not specified	Random number seed: using the same seed improves generation stability (results are not guaranteed to be identical)	Random value

⚠️ Usage Notes

🚫 Content Policy: Generation of illegal, vulgar, infringing, or other non-compliant content is prohibited. All requests are subject to content review; non-compliant requests will be rejected and recorded.

💾 Resource Retention: Generated image files will be retained for 7 days and then automatically deleted. Please download and save them promptly.

🔧 Debugging Recommendation: New users are advised to use the online debugging tool to quickly test parameter configurations.

Pricing:0.08 PTC/image

Request

Header Params

Body Params application/json

Example

{
    "model": "qwen-image-max",
    "input": {
        "messages": [
            {
                "role": "user",
                "content": [
                    {
                        "text": "An elegant and solemn couplet hangs in the hall. The room is decorated in a quiet and classic Chinese style, with some blue and white porcelain placed on the table. On the couplet, the left side reads “Righteousness is innate; humans and machines follow the same path, and good thinking leads to new ideas”, and the right side reads “Connecting to the clouds and endowing wisdom, the universe unfolds numbers with lofty aspirations”. The horizontal scroll reads “Wisdom Initiates Tongyi”. The calligraphy is elegant and flowing. In the middle hangs a Chinese-style painting depicting the Yueyang Tower."
                    }
                ]
            }
        ]
    },
    "parameters": {
        "negative_prompt": "Low resolution, low image quality, limb deformities, finger deformities, over-saturated images, a wax figure-like appearance, lack of facial details, excessive smoothness, and an AI-generated look. Chaotic composition. Blurred and distorted text.",
        "prompt_extend": true,
        "watermark": false,
        "size": "1328*1328"
    }
}

Request Code Samples

Shell

JavaScript

Java

Swift

PHP

Python

HTTP

Objective-C

Ruby

OCaml

Dart

curl --location --request POST 'https://api.302.ai/aliyun/api/v1/services/aigc/multimodal-generation/generation' \
--header 'Authorization: Bearer ' \
--header 'Content-Type: application/json' \
--data-raw '{
    "model": "qwen-image-max",
    "input": {
        "messages": [
            {
                "role": "user",
                "content": [
                    {
                        "text": "An elegant and solemn couplet hangs in the hall. The room is decorated in a quiet and classic Chinese style, with some blue and white porcelain placed on the table. On the couplet, the left side reads “Righteousness is innate; humans and machines follow the same path, and good thinking leads to new ideas”, and the right side reads “Connecting to the clouds and endowing wisdom, the universe unfolds numbers with lofty aspirations”. The horizontal scroll reads “Wisdom Initiates Tongyi”. The calligraphy is elegant and flowing. In the middle hangs a Chinese-style painting depicting the Yueyang Tower."
                    }
                ]
            }
        ]
    },
    "parameters": {
        "negative_prompt": "Low resolution, low image quality, limb deformities, finger deformities, over-saturated images, a wax figure-like appearance, lack of facial details, excessive smoothness, and an AI-generated look. Chaotic composition. Blurred and distorted text.",
        "prompt_extend": true,
        "watermark": false,
        "size": "1328*1328"
    }
}'

Responses

🟢200success

application/json

Body

Example

{
    "output": {
        "choices": [
            {
                "finish_reason": "stop",
                "message": {
                    "role": "assistant",
                    "content": [
                        {
                            "image": "https://file.302.ai/gpt/imgs/20250821/b7a6dce523e947859f0d1cc3c985f099.png"
                        }
                    ]
                }
            }
        ]
    },
    "usage": {
        "width": 1248,
        "image_count": 1,
        "height": 832
    },
    "request_id": "83c1cd46-35aa-9c2a-9e41-ac788b8fcfdc"
}

Modified at 2026-01-05 06:07:02

Qwen-Image (Alibaba Cloud Deployment)

T2I (Text-to-Image Alibaba Cloud)