> ## Documentation Index
> Fetch the complete documentation index at: https://docs.myrouter.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Seedream Text-to-Image 3.0

Seedream 3.0 is an advanced text-to-image model that can efficiently and quickly generate high-quality images from text prompts.

<Tip>
  Currently, only the model version `seedream-3-0-t2i-250415` is supported for Seedream 3.0.
</Tip>

## Request Headers

<ParamField header="Content-Type" type="string" required={true}>
  Enum: `application/json`
</ParamField>

<ParamField header="Authorization" type="string" required={true}>
  Bearer authentication format: Bearer \{\{API Key}}.
</ParamField>

## Request Body

<ParamField body="prompt" type="string" required={true}>
  Text prompt for image generation.
</ParamField>

<ParamField body="model" type="string" required={false}>
  The model ID or inference Endpoint ID used for this request. Currently only `seedream-3-0-t2i-250415` is supported for Seedream 3.0.
</ParamField>

<ParamField body="response_format" type="string" required={false}>
  Specifies the image format returned in the response. Default: `url`.<br />
  Supported values:<br />

  * `"url"`: Returns a downloadable JPEG image link.
  * `"b64_json"`: Returns a base64-encoded image JSON string.
</ParamField>

<ParamField body="size" type="string" required={false}>
  Specifies the size of the generated image (pixels, width x height), range is \[512x512, 2048x2048]. Default: `1024x1024`.<br />
  Recommended aspect ratios and resolutions:<br />

  * `1024x1024` (1:1)
  * `864x1152` (3:4)
  * `1152x864` (4:3)
  * `1280x720` (16:9)
  * `720x1280` (9:16)
  * `832x1248` (2:3)
  * `1248x832` (3:2)
  * `1512x648` (21:9)
</ParamField>

<ParamField body="seed" type="integer" required={false}>
  Seed value for controlling randomness in image generation. Range: \[-1, 2147483647]. If not specified, one will be automatically generated. To reproduce the same result, use the same seed. Default: `-1`.
</ParamField>

<ParamField body="guidance_scale" type="number" required={false}>
  Controls how closely the output image matches the input prompt. Higher values give the model less freedom and produce stronger correlation with the prompt. Range: \[1, 10]. Default: `2.5`.
</ParamField>

<ParamField body="watermark" type="boolean" required={false}>
  Whether to add a watermark to the generated image. Default: `true`.<br />

  * `false`: No watermark added
  * `true`: Adds an "AI generated" watermark in the bottom-right corner of the image
</ParamField>

## Response

<ResponseField name="image_urls" type="string[]" required={false}>
  Array of generated image links. When `response_format` is set to `"url"`, this array contains downloadable links for the generated images.
</ResponseField>

<ResponseField name="binary_data_base64" type="string[]" required={false}>
  Array of Base64-encoded image data. When `response_format` is set to `"b64_json"`, this array contains base64-encoded image JSON strings.
</ResponseField>