Legacy Wan Text-to-Image

Legacy Alibaba Wanxiang text-to-image models including wan2.6-t2i, wan2.2, and wanx2.1 series, offering multiple sizes, batch generation, and negative prompt support.

Try it Online

Test in Playground ↗ · View Models ↗

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="wan2.6-t2i",
    prompt="An orange tabby cat dozing under a cherry blossom tree",
    size="1280x1280",
    n=2
)

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

Supported Models

Model	Category	Features	Max Resolution	Sync
`wan2.6-t2i`	2.6 Latest	Latest protocol, sync+async, supports prompt_extend	1440²	Yes
`wan2.2-t2i-plus`	2.2 High-Qual	High quality legacy, each side [512, 1440]	1440²	No (async)
`wan2.2-t2i-flash`	2.2 Fast	Fast legacy, high speed and low cost	1440²	No (async)
`wanx2.1-t2i-turbo`	2.1 Fast	2.1 fast variant	1440²	No (async)
`wanx2.1-t2i-plus`	2.1 High-Qual	2.1 high quality variant	1440²	No (async)

Model Comparison

Feature	wan2.6-t2i	wan2.2-t2i-plus	wan2.2-t2i-flash	wanx2.1-t2i-turbo	wanx2.1-t2i-plus
Sync Response	Yes	No	No	No	No
prompt_extend	Yes	No	No	No	No
watermark	Yes	No	No	No	No
negative_prompt	Yes	Yes	Yes	Yes	Yes
Batch Gen (n)	1–4	1–4	1–4	1–4	1–4
Max Resolution	1440²	1440²	1440²	1440²	1440²

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
n	integer	No	`4`	Number of images to generate, range 1–4
response_format	string	No	`url`	Response format: `url` or `b64_json`
negative_prompt	string	No	—	Negative prompt describing unwanted content
seed	integer	No	—	Random seed for reproducible results

Supported Sizes

wan2.6-t2i

Total pixel range: [1280², 1440²]

Size	Aspect Ratio	Description
`1280x1280`	1:1	Square
`1696x960`	16:9	Widescreen
`960x1696`	9:16	Full-screen tall
`1472x1104`	4:3	Landscape
`1104x1472`	3:4	Portrait

wan2.2 / wanx2.1 Series

Each side range: [512, 1440]

Size	Aspect Ratio	Description
`1024x1024`	1:1	Square
`1440x1440`	1:1	Square HD
`1280x720`	16:9	Widescreen
`720x1280`	9:16	Full-screen tall
`1024x1440`	~3:4	Portrait
`1440x1024`	~4:3	Landscape
`768x1152`	2:3	Portrait
`1152x768`	3:2	Landscape
`960x1280`	3:4	Portrait
`1280x960`	4:3	Landscape
`512x1024`	1:2	Narrow portrait
`1024x512`	2:1	Wide landscape

Extended Parameters

wan2.6-t2i Only

Field	Type	Required	Default	Description
prompt_extend	boolean	No	`true`	Whether to automatically extend the prompt for better results
watermark	boolean	No	`false`	Whether to add a watermark. Set to `true` to enable

Note

The wan2.2 and wanx2.1 legacy models do not support prompt_extend or watermark parameters in compatible mode.

Common Extended Parameters (All Models)

Field	Type	Description
negative_prompt	string	Negative prompt describing unwanted content
seed	integer	Random seed for reproducible results

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": "wan2.6-t2i",
    "prompt": "An orange tabby cat dozing under a cherry blossom tree",
    "size": "1280x1280",
    "n": 2
  }'

python

from openai import OpenAI

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

response = client.images.generate(
    model="wan2.6-t2i",
    prompt="An orange tabby cat dozing under a cherry blossom tree",
    size="1280x1280",
    n=2
)

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: 'wan2.6-t2i',
  prompt: 'An orange tabby cat dozing under a cherry blossom tree',
  size: '1280x1280',
  n: 2,
})

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

Using Negative Prompts

bash

curl https://open.dieyuyun.com/v1/images/generations \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "wan2.6-t2i",
    "prompt": "An orange tabby cat dozing under a cherry blossom tree, watercolor style",
    "size": "1280x1280",
    "negative_prompt": "blurry, low quality, deformed",
    "n": 2
  }'

Disable Prompt Extension (wan2.6-t2i)

bash

curl https://open.dieyuyun.com/v1/images/generations \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "wan2.6-t2i",
    "prompt": "An orange tabby cat dozing under a cherry blossom tree",
    "size": "1280x1280",
    "prompt_extend": false,
    "n": 2
  }'

Response Format

Success Response

json

{
  "created": 1717000000,
  "data": [
    {
      "url": "https://...",
      "revised_prompt": "An orange tabby cat peacefully dozing beneath a blooming cherry blossom tree, petals gently falling, dappled sunlight filtering through the leaves"
    },
    {
      "url": "https://...",
      "revised_prompt": "An orange tabby cat peacefully dozing beneath a blooming cherry blossom tree, petals gently falling, dappled sunlight filtering through the leaves"
    }
  ]
}

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

Sync generation: Use wan2.6-t2i when you need synchronous results — it is the only model that supports sync responses, eliminating the need to poll for task_id.
High quality: For the best image quality, choose wan2.2-t2i-plus or wanx2.1-t2i-plus, which excel in detail and composition.
Fast generation: For speed and low cost, choose wan2.2-t2i-flash or wanx2.1-t2i-turbo, ideal for rapid iteration.
Batch generation: All models support generating 1–4 images per request (n defaults to 4), useful for quick comparison and design selection.
Negative prompts: Use negative_prompt to exclude unwanted elements, e.g. negative_prompt: "blurry, low quality, deformed".
Prompt extension: wan2.6-t2i has prompt_extend enabled by default (true), which automatically enhances your prompt. Set to false for precise control.
Size selection: Choose the right size for your use case — 9:16 for social media stories, 16:9 for banners, 1:1 for general purpose.
Async model note: The wan2.2 and wanx2.1 series return asynchronously. You need to poll by task_id to retrieve results.

Rate Limits

See Rate Limits for details.

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

Legacy Wan Text-to-Image ​

Quick Start ​

Endpoint ​

Authentication ​

Supported Models ​

Model Comparison ​

Standard Request Fields ​

Common to All Models ​

Supported Sizes ​

wan2.6-t2i ​

wan2.2 / wanx2.1 Series ​

Extended Parameters ​

wan2.6-t2i Only ​

Common Extended Parameters (All Models) ​

Request Examples ​

Using Negative Prompts ​

Disable Prompt Extension (wan2.6-t2i) ​

Response Format ​

Success Response ​

Error Response ​

Best Practices ​

Rate Limits ​

Related Documentation ​

Legacy Wan Text-to-Image

Quick Start

Endpoint

Authentication

Supported Models

Model Comparison

Standard Request Fields

Common to All Models

Supported Sizes

wan2.6-t2i

wan2.2 / wanx2.1 Series

Extended Parameters

wan2.6-t2i Only

Common Extended Parameters (All Models)

Request Examples

Using Negative Prompts

Disable Prompt Extension (wan2.6-t2i)

Response Format

Success Response

Error Response

Best Practices

Rate Limits

Related Documentation