Wan2.6-Image (Image Generation and Editing)

Tongyi Wanxiang image generation model supports image editing and mixed text-image output, meeting diverse generation and integration needs.

Usage Instructions

Parameter Requirements

Depending on the value of the enable_interleave (image generation mode) parameter, the required parameters/rules to be followed vary:

Text-image interleaving mode (enable_interleave: true)

Core rules:

Length of image array: 0~1 (images can be omitted)

n (number of generated images): fixed at 1

max_images: effective, value range 1~5 (default 5)

Required parameters: model, input.messages (must contain the text field)

Optional parameters: negative_prompt, watermark, seed, size

Image editing mode (enable_interleave: false, default)

Core rules:

Length of image array: 1~3 (at least 1 image must be provided)

n (number of generated images): 1~4 (default 4)

prompt_extend: effective, intelligent prompt rewriting is enabled by default

Required parameters: model, input.messages (must contain both text and image fields)

Optional parameters: negative_prompt, prompt_extend, watermark, size, seed

Character Length Limits

Parameter Name	Length Limit	Handling of Exceeding Limits
`input.messages.content.text`	2000 characters (Chinese and English)	Automatically truncated
`parameters.negative_prompt`	500 characters (Chinese and English)	Automatically truncated

Note: Each character (Chinese character/letter/number/symbol) counts as 1 character; only one text field is supported. Passing multiple or none will result in an error.

Image Format Limits

Limit Type	Specific Rules
Format	JPEG, JPG, PNG (without transparent channel), BMP, WEBP
Resolution	Width/height range [384, 5000] pixels
File Size	No more than 10MB
Input Format	1. Publicly accessible URL (HTTP/HTTPS) 2. Base64 encoded string (format: `data:{MIME_type};base64,{base64_data}`)

Notes

⚠️ During the testing phase, it is recommended to set n to 1 to reduce verification costs

📌 When enable_interleave=true, max_images is the upper limit of the number of generated images, and the actual number may be less than the set value

✅ The seed value range is [0, 2147483647]. The same seed can improve generation stability (not completely consistent)

🔖 Watermark is disabled by default. When enabled, the text "AI Generated" will be displayed in the lower right corner of the image

⏱️ Enabling prompt_extend will increase processing time by 3-4 seconds and only takes effect in image editing mode

Price: 0.03 PTC per image

Request

Header Params

Body Params application/json

Examples

Refer to image-to-imageMixed text and image output

{
    "model": "wan2.6-image",
    "input": {
        "messages": [
            {
                "role": "user",
                "content": [
                    {
                        "text": "Replace the floral dress with a vintage-style lace long dress, featuring exquisite embroidery details on the neckline and cuffs.。"
                    },
                    {
                        "image": "https://help-static-aliyun-doc.aliyuncs.com/assets/img/zh-CN/3943878571/p1011665.webp"
                    }
                ]
            }
        ]
    },
    "parameters": {
        "negative_prompt": "",
        "prompt_extend": true,
        "watermark": false,
        "n": 2,
        "size": "1280*1280"
    }
}

Request Code Samples

Shell

JavaScript

Java

Swift

PHP

Python

HTTP

Objective-C

Ruby

OCaml

Dart

application/json

Body

Example

{
    "request_id": "24d5a019-e4ed-47de-a8c3-0badbd17a084",
    "output": {
        "task_id": "2132eabd-ad49-4ef5-be52-34b1a690fcc5",
        "task_status": "PENDING"
    }
}

Wan2.6-Image (Image Generation and Editing)

Usage Instructions#

Parameter Requirements#

Character Length Limits#

Image Format Limits#

Notes#