Wan 2.6 Image (International)

wan26_image_intl

wan26_image_intl Wan 2.6 Image International Qwen image editing text to image image to image multi image reference international

Wan 2.6 Image is an Alibaba image generation and editing model. It supports prompt-only image generation, image editing, and interleaved text-image output through the international (Singapore) node.

Эндпоинты

POST /v1/queue/wan26_image_intl # Создать задачу

GET /v1/queue/wan26_image_intl/requests/{request_id}/status # Проверить статус

GET /v1/queue/wan26_image_intl/requests/{request_id}/response # Получить результат

PUT /v1/queue/wan26_image_intl/requests/{request_id}/cancel # Отменить задачу

Tip

Добавляйте заголовок аутентификации в каждый запрос: Authorization: Bearer YOUR_API_KEY.

Mode selection

For prompt-only image generation or interleaved text-image output, set enable_interleave to true, keep n as 1, and use max_images to control the output limit.

For image editing, set enable_interleave to false and provide at least one input image with image, image_url, or another image slot.

Параметры запроса

curl --location "https://odirouter.ai/model/v1/queue/wan26_image_intl" \
  --header "Authorization: Bearer YOUR_API_KEY" \
  --header "Content-Type: application/json" \
  --data '{
    "prompt": "Turn the input image into a clean studio product photo on a white background.",
    "image_url": "https://example.com/input.webp",
    "enable_interleave": false,
    "size": "1K",
    "n": 1,
    "prompt_extend": false,
    "watermark": false
  }'

role string

Message role. Use user.

prompt string обязательно

Positive prompt describing the desired generated or edited image content, style, and composition. Chinese and English are supported, and content beyond the model limit is truncated automatically.

Maximum length: 2000 characters

image string

Input image URL or Base64-encoded string.

Supported formats: JPEG, JPG, PNG, BMP, WEBP. PNG images do not support transparent channels.

Image width and height: 240 - 8000 pixels

Maximum file size: 10MB

image_url string

Input image URL or Base64-encoded string.

Supported formats: JPEG, JPG, PNG, BMP, WEBP. PNG images do not support transparent channels.

Image width and height: 240 - 8000 pixels

Maximum file size: 10MB

image_2 string

Input image URL or Base64-encoded string.

Supported formats: JPEG, JPG, PNG, BMP, WEBP. PNG images do not support transparent channels.

Image width and height: 240 - 8000 pixels

Maximum file size: 10MB

image_3 string

Input image URL or Base64-encoded string.

Supported formats: JPEG, JPG, PNG, BMP, WEBP. PNG images do not support transparent channels.

Image width and height: 240 - 8000 pixels

Maximum file size: 10MB

image_4 string

Input image URL or Base64-encoded string.

Supported formats: JPEG, JPG, PNG, BMP, WEBP. PNG images do not support transparent channels.

Image width and height: 240 - 8000 pixels

Maximum file size: 10MB

enable_interleave bool

Controls image generation mode.

false: image editing mode, supporting 1-4 reference images. Provide at least one input image with image, image_url, image_2, image_3, or image_4.

true: interleaved text-image output mode, supporting prompt-only generation or one input image.

Default: false

negative_prompt string

Negative prompt describing content that should not appear in the image.

Maximum length: 500 characters

size string

Output image resolution. Image editing mode can use reference tiers 1K or 2K, or explicit pixel dimensions.

In image editing mode, total pixels are approximately from 768x768 to 2048x2048, with aspect ratio from 1:4 to 4:1.

In interleaved text-image output mode, explicit pixel dimensions support total pixels approximately from 768x768 to 1280x1280.

Default: 1K

n int

Number of images to generate. This parameter directly affects cost; set it to 1 when you need a single output.

Image editing mode range: 1 - 4

In interleaved mode, this must be 1. Use max_images to control the output limit.

Default: 1

max_images int

Maximum number of generated images in interleaved text-image output mode.

Use with enable_interleave=true. In this mode, keep n set to 1.

Range: 1 - 4

Default: 1

prompt_extend bool

Whether to enable prompt rewriting. When enabled, the positive prompt is optimized while the negative prompt is unchanged.

Options: true, false

Default: true

watermark bool

Whether to add a watermark in the lower-right corner of the image.

Options: false, true

Default: false

seed int

Random seed. When omitted, the system generates a random seed automatically. A fixed seed can improve reproducibility, but identical results are not guaranteed.

Range: 0 - 2147483647

Параметры ответа

{
  "request_id": "mmsu_01hxyz...",
  "response_url": "https://odirouter.ai/model/v1/queue/wan26_image_intl/requests/mmsu_01hxyz.../response",
  "status_url": "https://odirouter.ai/model/v1/queue/wan26_image_intl/requests/mmsu_01hxyz.../status",
  "cancel_url": "https://odirouter.ai/model/v1/queue/wan26_image_intl/requests/mmsu_01hxyz.../cancel",
  "queue_position": 0
}

Поле	Тип	Описание
request_id	string	Идентификатор задачи для последующей проверки статуса, получения результата и отмены
response_url	string	URL для получения итогового результата
status_url	string	URL для проверки статуса задачи
cancel_url	string	URL для отмены задачи
queue_position	int	Текущая позиция в очереди

Следующие операции

Проверить статус

Получить результат

Отменить задачу

Обработка ошибок