Seedream 4.5

POST /v3/seedream-4.5

Seedream 4.5

cURL

curl --request POST \
  --url https://api.myrouter.ai/v3/seedream-4.5 \
  --header 'Authorization: <authorization>' \
  --header 'Content-Type: <content-type>' \
  --data '
{
  "size": "<string>",
  "image": [
    {}
  ],
  "prompt": "<string>",
  "watermark": true,
  "optimize_prompt_options": {
    "mode": "<string>"
  },
  "sequential_image_generation": "<string>",
  "sequential_image_generation_options": {
    "max_images": 123
  }
}
'

200

{
  "images": [
    {}
  ]
}

Generate images based on input text prompts and/or reference images. Supports generating single images or image sequences (a set of content-related images).

Request Headers

Content-Type · string · required

Enum: application/json

Authorization · string · required

Bearer authentication format: Bearer {{API Key}}.

Request Body

size · string · default: "2048x2048"

Specifies the dimensions of the generated image. Method 1: Specify resolution (2K, 4K). Method 2: Specify width and height in pixels (e.g., 2048x2048). Total pixel range: [3686400, 16777216]. Aspect ratio range: [1/16, 16].

image · array

Array of input image data, supporting URL or Base64 encoding. Up to 14 reference images supported. Supported image formats: jpeg, png, webp, bmp, tiff, gif.

Array length: 1 - 14

prompt · string · required

Text prompt for image generation. Supports both Chinese and English. Recommended no more than 300 Chinese characters or 600 English words.

watermark · boolean · default: true

Whether to add a watermark to the generated image.

optimize_prompt_options · object

Configuration for the prompt optimization feature.

Hide properties

mode · string · default: "standard"

Sets the mode for the prompt optimization feature. standard: standard mode with higher quality but longer processing time. fast: fast mode with shorter processing time but average quality. Currently only standard mode is supported.

Possible values: standard

sequential_image_generation · string · default: "disabled"

Controls whether to enable image sequence generation. auto: automatic mode where the model determines whether to return an image sequence based on the prompt. disabled: disables image sequence generation, producing only a single image.

Possible values: auto, disabled

sequential_image_generation_options · object

Configuration for the image sequence generation feature. Only takes effect when sequential_image_generation is set to auto.

Hide properties

max_images · integer · default: 15

Specifies the maximum number of images that can be generated per request. The number of input reference images + the number of generated images must be no more than 15.

Range: [1, 15]

Response

images · array

Array of generated image data.