Qwen-Image Generation

Qwen-Image series image generation models with excellent Chinese language understanding and text rendering capabilities, supporting multiple sizes, batch generation, and watermark control.

Try it Online

Test in Playground ↗ · View Models ↗

Supported Models

Model	Features	Max Resolution	n (Batch Size)	Response
`qwen-image-2.0-pro`	Text rendering, up to 2K resolution	2K (2368)	1–6	Sync
`qwen-image-2.0`	Accelerated 2.0-pro, same parameters	2K (2368)	1–6	Sync
`qwen-image-max`	Fixed preset sizes, high quality	1664	Fixed 1	Sync
`qwen-image-plus`	Fixed preset sizes, async task support	1664	Fixed 1	Sync/Async
`qwen-image`	Basic model, async task support	1536	Fixed 1	Sync/Async

Model Comparison

Feature	qwen-image-2.0-pro	qwen-image-2.0	qwen-image-max	qwen-image-plus	qwen-image
Text Rendering	Yes	Yes	No	No	No
Max Resolution	2368	2368	1664	1664	1536
Batch Gen (n)	1–6	1–6	1	1	1
negative_prompt	Yes	Yes	Yes	Yes	Yes
prompt_extend	Yes	Yes	Yes	Yes	Yes
Async Tasks	No	No	No	Yes	Yes

Quick Start

python

from openai import OpenAI

client = OpenAI(
    base_url="https://open.dieyuyun.com/v1",
    api_key="YOUR_API_KEY"
)

response = client.images.generate(
    model="qwen-image-2.0-pro",
    prompt="A traditional Chinese landscape painting, misty mountains with a small bridge over flowing water",
    size="2048x2048",
    n=1
)

print(response.data[0].url)

Endpoint

Item	Value
Method	POST
Path	/v1/images/generations
Base URL	`https://open.dieyuyun.com`
Protocol	OpenAI Images API

Authentication

Bearer Token: Authorization: Bearer YOUR_API_KEY

Standard Request Fields

Common to All Models

Field	Type	Required	Default	Description
model	string	Yes	—	Model identifier. See Supported Models
prompt	string	Yes	—	Image description text
size	string	No	See size tables	Image size. Supported sizes vary by model
response_format	string	No	`url`	Response format: `url` or `b64_json`
watermark	boolean	No	`true`	Whether to add a watermark. Set to `false` to disable
seed	integer	No	—	Random seed for reproducible results
negative_prompt	string	No	—	Negative prompt describing unwanted content
prompt_extend	boolean	No	`true`	Whether to automatically extend the prompt for better results

2.0 Series Models (qwen-image-2.0-pro / qwen-image-2.0)

Field	Type	Description
n	integer	Number of images to generate, range 1–6
size	string	Supports the following sizes (up to 2K):

Size	Aspect Ratio	Description
`2048x2048`	1:1	Square
`2688x1536`	16:9	Widescreen
`1536x2688`	9:16	Full-screen tall
`2368x1728`	4:3	Landscape
`1728x2368`	3:4	Portrait

max / plus / Basic Models (qwen-image-max / qwen-image-plus / qwen-image)

Field	Type	Description
n	integer	Fixed at 1
size	string	Supports the following presets:

Size	Aspect Ratio	Description
`1664x928`	16:9	Widescreen
`1472x1104`	4:3	Landscape
`1328x1328`	1:1	Square
`1104x1472`	3:4	Portrait
`928x1664`	9:16	Full-screen tall

Extended Parameters

Qwen-Image supports additional parameters via the parameters object:

Field	Type	Required	Default	Description
parameters.watermark	boolean	No	`true`	Whether to add a watermark. Set to `false` to disable

json

{
  "model": "qwen-image-2.0-pro",
  "prompt": "A beautiful sunset over mountains",
  "size": "2048x2048",
  "parameters": {
    "watermark": false
  }
}

Request Examples

cURLPythonNode.js

bash

curl https://open.dieyuyun.com/v1/images/generations \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "qwen-image-2.0-pro",
    "prompt": "A traditional Chinese landscape painting, misty mountains with a small bridge over flowing water",
    "size": "2048x2048",
    "n": 1
  }'

python

from openai import OpenAI

client = OpenAI(
    base_url="https://open.dieyuyun.com/v1",
    api_key="YOUR_API_KEY"
)

response = client.images.generate(
    model="qwen-image-2.0-pro",
    prompt="A traditional Chinese landscape painting, misty mountains with a small bridge over flowing water",
    size="2048x2048",
    n=1
)

print(response.data[0].url)

javascript

import OpenAI from 'openai'

const client = new OpenAI({
  baseURL: 'https://open.dieyuyun.com/v1',
  apiKey: 'YOUR_API_KEY',
})

const response = await client.images.generate({
  model: 'qwen-image-2.0-pro',
  prompt: 'A traditional Chinese landscape painting, misty mountains with a small bridge over flowing water',
  size: '2048x2048',
  n: 1,
})

console.log(response.data[0].url)

Batch Generation Example (2.0 Series)

bash

curl https://open.dieyuyun.com/v1/images/generations \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "qwen-image-2.0-pro",
    "prompt": "An orange tabby cat sitting on a windowsill in warm sunlight, watercolor style",
    "size": "2048x2048",
    "n": 4
  }'

Disable Watermark Example

bash

curl https://open.dieyuyun.com/v1/images/generations \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "qwen-image-2.0-pro",
    "prompt": "A traditional Chinese landscape painting, misty mountains with a small bridge over flowing water",
    "size": "2048x2048",
    "parameters": {
      "watermark": false
    },
    "n": 1
  }'

Response Format

Success Response

json

{
  "created": 1717000000,
  "data": [
    {
      "url": "https://...",
      "revised_prompt": "A traditional Chinese landscape painting with layered misty mountains rendered in ink wash style, a small stone bridge over a gentle stream, a thatched cottage nestled among trees"
    }
  ]
}

Field	Type	Description
created	integer	Unix timestamp
data	array	Array of generated results
data[].url	string	Image URL
data[].revised_prompt	string	The revised/optimized prompt used for generation

Error Response

See Error Codes for details.

Best Practices

Chinese prompts: Qwen-Image excels at understanding Chinese prompts, making it ideal for Chinese-style art, traditional painting, and culturally specific scenes.
Text rendering: When generating images that include text, prefer qwen-image-2.0-pro or qwen-image-2.0 — they are optimized for rendering text within images.
Batch generation: The 2.0 series supports generating up to 6 images per request (n up to 6), ideal for rapid iteration and design comparison.
Choose the right model: For quality, use qwen-image-2.0-pro. For speed, use qwen-image-2.0. For other use cases, choose max, plus, or the basic model.
Disable watermark: For commercial use, set parameters.watermark: false to generate watermark-free images.
Negative prompts: Use negative_prompt to exclude unwanted elements, e.g. negative_prompt: "blurry, low quality".
Prompt extension: prompt_extend defaults to true — the system automatically enhances your prompt. Set to false for precise control.
Synchronous response: The 2.0 series and max models return results synchronously. No task_id polling is needed.
Style diversity: Supports realistic, illustration, traditional painting, oil painting, and many other styles. Specify the style explicitly in your prompt for best results.

Rate Limits

See Rate Limits for details.

Wanxiang 2.7 - Tongyi Wanxiang video generation model
Z-Image Turbo - High-speed image generation model
GPT Image 2 - OpenAI image generation model
Model List & Pricing - Browse available models and pricing
Playground - Test image generation interactively

Qwen-Image Generation ​

Supported Models ​

Model Comparison ​

Quick Start ​

Endpoint ​

Authentication ​

Standard Request Fields ​

Common to All Models ​

2.0 Series Models (qwen-image-2.0-pro / qwen-image-2.0) ​

max / plus / Basic Models (qwen-image-max / qwen-image-plus / qwen-image) ​

Extended Parameters ​

Request Examples ​

Batch Generation Example (2.0 Series) ​

Disable Watermark Example ​

Response Format ​

Success Response ​

Error Response ​

Best Practices ​

Rate Limits ​

Related Documentation ​

Qwen-Image Generation

Supported Models

Model Comparison

Quick Start

Endpoint

Authentication

Standard Request Fields

Common to All Models

2.0 Series Models (qwen-image-2.0-pro / qwen-image-2.0)

max / plus / Basic Models (qwen-image-max / qwen-image-plus / qwen-image)

Extended Parameters

Request Examples

Batch Generation Example (2.0 Series)

Disable Watermark Example

Response Format

Success Response

Error Response

Best Practices

Rate Limits

Related Documentation