seedream-5.0-pro

Use doubao-seedream-5-0-pro-260628 with the synchronous image generation endpoint. Select a capability tab to view its request and response shape.

This model does not support grouped image generation, web search, or streaming. Do not send sequential_image_generation, sequential_image_generation_options, tools, or stream.

Text to image
Image editing
Transparent background
Layer decomposition

Create an image

POST /v1/images/generations

cURL

curl --request POST \
  --url https://www.anyfast.ai/v1/images/generations \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '{
    "model": "doubao-seedream-5-0-pro-260628",
    "prompt": "A premium product photograph of a white ceramic cup on a light gray background, centered composition, soft shadow, no text",
    "size": "1.5K",
    "response_format": "url",
    "output_format": "jpeg",
    "watermark": false,
    "optimize_prompt_options": { "mode": "fast" }
  }'

string

required

Must be doubao-seedream-5-0-pro-260628.

string

required

Image-generation prompt. Chinese prompts should generally stay within 300 characters and English prompts within 600 words.

string

default:"2K"

1K, 1.5K, 2K, or a valid explicit widthxheight value. Explicit dimensions require 921,600–4,624,220 total pixels and an aspect ratio in [1/16, 16].

string

default:"url"

url or b64_json. URL results expire after 24 hours.

string

default:"jpeg"

jpeg or png.

boolean

default:true

Whether to add the AI-generated watermark.

object

Prompt optimization configuration.

string

default:"standard"

standard or fast. Both values are supported.

Create response

200

{
  "model": "doubao-seedream-5-0-pro-260628",
  "created": 1787562068,
  "data": [
    {
      "url": "https://example.com/generated-image.jpeg",
      "size": "1536x1536",
      "output_format": "jpeg"
    }
  ],
  "usage": {
    "input_images": 0,
    "generated_images": 1,
    "output_tokens": 9216,
    "total_tokens": 9216
  }
}

Edit or combine images

POST /v1/images/generations

cURL

curl --request POST \
  --url https://www.anyfast.ai/v1/images/generations \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '{
    "model": "doubao-seedream-5-0-pro-260628",
    "prompt": "Use the composition of image 1 and the visual style of image 2.",
    "image": [
      "https://example.com/composition.png",
      "https://example.com/style.png"
    ],
    "size": "1K",
    "response_format": "url",
    "output_format": "jpeg",
    "watermark": false
  }'

string

required

Must be doubao-seedream-5-0-pro-260628.

string

required

Editing or reference-generation instruction. Coordinate edits can use normalized <point> and <bbox> tags.

string | string[]

required

Image URL, Base64 data URI, or an array of up to 10 references.

string

default:"2K"

1K, 1.5K, 2K, or a valid explicit pixel size.

string

default:"url"

url or b64_json.

string

default:"jpeg"

jpeg or png.

boolean

default:true

Whether to add the AI-generated watermark.

string

default:"standard"

standard or fast.

Create response

200

{
  "model": "doubao-seedream-5-0-pro-260628",
  "created": 1787562132,
  "data": [
    {
      "url": "https://example.com/fused-image.jpeg",
      "size": "1024x1024",
      "output_format": "jpeg"
    }
  ],
  "usage": {
    "input_images": 2,
    "generated_images": 1,
    "output_tokens": 4096,
    "total_tokens": 4096
  }
}

Normal reference images support jpeg, png, webp, bmp, tiff, gif, heic, and heif. Each image must be smaller than 30 MB, use an aspect ratio in [1/16, 16], have width and height greater than 14 px, and contain no more than 36,000,000 pixels.

Preserve transparency

POST /v1/images/generations

cURL

curl --request POST \
  --url https://www.anyfast.ai/v1/images/generations \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '{
    "model": "doubao-seedream-5-0-pro-260628",
    "prompt": "Keep the subject and refine the edges while preserving the transparent background.",
    "image": "https://example.com/transparent-source.png",
    "background": "transparent",
    "size": "1K",
    "response_format": "url",
    "output_format": "png",
    "watermark": false
  }'

string

required

Must be doubao-seedream-5-0-pro-260628.

string

required

Editing instruction.

string

required

Exactly one input image with an alpha channel. PNG is supported; JPEG cannot carry transparency.

string

default:"opaque"

Use transparent to preserve transparent output, or opaque for a normal background.

string

default:"2K"

1K, 1.5K, 2K, or a valid explicit pixel size.

string

required

Must be png when background is transparent.

string

default:"url"

url or b64_json.

boolean

default:true

Whether to add the AI-generated watermark.

Create response

200

{
  "model": "doubao-seedream-5-0-pro-260628",
  "created": 1787562204,
  "data": [
    {
      "url": "https://example.com/transparent-result.png",
      "size": "1024x1024",
      "output_format": "png"
    }
  ],
  "usage": {
    "input_images": 1,
    "generated_images": 1,
    "output_tokens": 4096,
    "total_tokens": 4096
  }
}

Decompose an image

POST /v1/images/generations

cURL

curl --request POST \
  --url https://www.anyfast.ai/v1/images/generations \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '{
    "model": "doubao-seedream-5-0-pro-260628",
    "prompt": "Separate the parrot subject from the background into one editable layer.",
    "image": "https://example.com/parrot-poster.png",
    "layer_decomposition": true,
    "size": "2K",
    "response_format": "url",
    "output_format": "jpeg",
    "watermark": false
  }'

string

required

Must be doubao-seedream-5-0-pro-260628.

string

Optional decomposition instruction. Use normalized <bbox> coordinates to target specific elements.

string

required

Exactly one PNG or JPEG image.

boolean

required

Set to true to return one background and up to 16 editable layers.

string

default:"auto"

auto, 1K, 1.5K, or 2K. Explicit pixel sizes are not supported in this mode.

string

default:"url"

url or b64_json.

string

default:"jpeg"

Controls only the background format. Extracted layers are always PNG.

boolean

default:true

Whether to add the AI-generated watermark.

Create response

200

{
  "model": "doubao-seedream-5-0-pro-260628",
  "created": 1787562257,
  "data": [
    {
      "url": "https://example.com/background.jpeg",
      "size": "2048x2048",
      "output_format": "jpeg",
      "z_index": 0
    },
    {
      "url": "https://example.com/parrot-layer.png",
      "size": "892x2179",
      "output_format": "png",
      "z_index": 1,
      "bounding_box": {
        "absolute": [730, 627, 1301, 2021],
        "normalized": [356, 306, 635, 986]
      },
      "name": "Parrot subject",
      "description": "The extracted parrot subject with a transparent background"
    }
  ],
  "usage": {
    "input_images": 1,
    "generated_images": 2,
    "output_tokens": 33768,
    "total_tokens": 33768
  }
}

Layer input must contain from 512 × 512 through 6000 × 6000 total pixels, use an aspect ratio in [1/16, 16], and be smaller than 30 MB. If any requested layer fails, the whole request fails.

Headers

string

required

Bearer authentication in the form Bearer <token>.

string

default:"application/json"

required

Request body format.

Response fields

string

Model used for the request. A provider-side normalized model name may be returned.

integer

Creation time as a Unix timestamp in seconds.

object[]

Generated images, or the background and extracted layers in decomposition mode.

string

Temporary image URL when response_format is url.

string

Base64 image data when response_format is b64_json.

string

Generated image format.

string

Generated image dimensions.

integer

Layer order in decomposition mode. The background uses 0.

string

Generated layer name in decomposition mode.

string

Generated layer description in decomposition mode.

integer[]

Layer bounds as [left, top, right, bottom] in output-image pixels.

integer[]

Layer bounds mapped to a 1000 × 1000 coordinate system.

integer

Number of generated outputs. In decomposition mode this includes the background and layers.

integer

Number of input images.

integer

Output token count.

integer

Total token count.

object

Request-level error information.

string

Request-level error code.

string

Request-level error message.

Error responses

Invalid or unsupported request parameters.

{
  "error": {
    "message": "The parameter is not supported by the current model.",
    "type": "BadRequest",
    "param": "sequential_image_generation",
    "code": "InvalidParameter"
  }
}

The API key is missing or invalid.

{
  "error": {
    "code": "unauthorized",
    "message": "Invalid API key"
  }
}

The request exceeded the applicable rate limit.

{
  "error": {
    "code": "rate_limit_exceeded",
    "message": "Rate limit exceeded"
  }
}

Overview

Native Compatible Endpoints

Model APIs

seedream-5.0-pro

Create an image

Create response

Edit or combine images

Create response

Preserve transparency

Create response

Decompose an image

Create response

Headers

Response fields

Error responses

​Create an image

​Create response

​Edit or combine images

​Create response

​Preserve transparency

​Create response

​Decompose an image

​Create response

​Headers

​Response fields

​Error responses

Create an image

Create response

Edit or combine images

Create response

Preserve transparency

Create response

Decompose an image

Create response

Headers

Response fields

Error responses