Doubao Seedream 5.0 Lite Image Generation

ByteDance's latest Seedream 5.0 Lite model, the first to support web-search-assisted image generation, with 2K/3K/4K multi-resolution and multi-image mode, delivering leading image quality and timeliness.

Try it Online

Test in Playground ↗ · View Models ↗

Quick Start

from openai import OpenAI

client = OpenAI(
    base_url="https://open.dieyuyun.com/v1",
    api_key="YOUR_API_KEY"
)

response = client.images.generate(
    model="doubao-seedream-5.0-lite",
    prompt="An orange tabby cat sitting on a windowsill watching the rain, watercolor style, warm tones",
    size="4096x4096",
    n=1
)

print(response.data[0].url)

Endpoint

Item	Value
Method	POST
Path	/v1/images/generations
Base URL	https://open.dieyuyun.com
Protocol	OpenAI Images API

Authentication

Bearer Token: Authorization: Bearer YOUR_API_KEY

Standard Request Fields

Field	Type	Required	Default	Description
model	string	Yes	—	Model identifier: doubao-seedream-5.0-lite
prompt	string	Yes	—	Image description text, supports Chinese and English, recommended under 300 Chinese characters
size	string	No	4096x4096	Image size (width x height in pixels)
n	integer	No	1	Number of images to generate (max 15). When >1, multi-image mode is enabled
response_format	string	No	url	Response format: url or b64_json

Supported Sizes

Doubao Seedream 5.0 Lite supports three resolution tiers: 2K, 3K (exclusive to 5.0 Lite), and 4K (default).

2K Resolution

Size	Aspect Ratio
2048x2048	1:1
2304x1728	4:3
1728x2304	3:4
2848x1600	16:9
1600x2848	9:16
2496x1664	3:2
1664x2496	2:3
3136x1344	21:9

3K Resolution (Exclusive to 5.0 Lite)

Size	Aspect Ratio
3072x3072	1:1
3456x2592	4:3
2592x3456	3:4
4096x2304	16:9
2304x4096	9:16
3744x2496	3:2
2496x3744	2:3
4704x2016	21:9

4K Resolution (Default)

Size	Aspect Ratio
4096x4096	1:1
4704x3520	4:3
3520x4704	3:4
5504x3040	16:9
3040x5504	9:16
4992x3328	3:2
3328x4992	2:3
6240x2656	21:9

Extended Parameters

Field	Type	Required	Default	Description
image	string/array	No	—	Reference image URLs or Base64, up to 14
watermark	boolean	No	false	Whether to add an AI-generated watermark
sequential_image_generation	string	No	disabled	auto for multi-image mode / disabled for single image
output_format	string	No	jpeg	Output image format: jpeg or png (4.5/4.0 always output jpeg)
tools_web_search	boolean	No	false	Enable web-search-assisted generation for up-to-date images (increases latency)

Request Examples

cURL

curl https://open.dieyuyun.com/v1/images/generations \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "doubao-seedream-5.0-lite",
    "prompt": "An orange tabby cat sitting on a windowsill watching the rain, watercolor style, warm tones",
    "size": "4096x4096",
    "n": 1
  }'

Python

from openai import OpenAI

client = OpenAI(
    base_url="https://open.dieyuyun.com/v1",
    api_key="YOUR_API_KEY"
)

response = client.images.generate(
    model="doubao-seedream-5.0-lite",
    prompt="An orange tabby cat sitting on a windowsill watching the rain, watercolor style, warm tones",
    size="4096x4096",
    n=1
)

print(response.data[0].url)

Node.js

import OpenAI from 'openai'

const client = new OpenAI({
  baseURL: 'https://open.dieyuyun.com/v1',
  apiKey: 'YOUR_API_KEY',
})

const response = await client.images.generate({
  model: 'doubao-seedream-5.0-lite',
  prompt: 'An orange tabby cat sitting on a windowsill watching the rain, watercolor style, warm tones',
  size: '4096x4096',
  n: 1,
})

console.log(response.data[0].url)

Response Format

Success Response

{
  "created": 1717000000,
  "model": "doubao-seedream-5.0-lite",
  "data": [
    {
      "url": "https://...",
      "size": "2048x2048"
    }
  ],
  "usage": {
    "generated_images": 1,
    "output_tokens": 16384,
    "total_tokens": 16384,
    "tool_usage": {
      "web_search": 0
    }
  }
}

Field	Type	Description
created	integer	Unix timestamp
model	string	Model identifier
data	array	Array of generated results
data[].url	string	Image URL
data[].size	string	Actual generated image dimensions
usage	object	Usage statistics
usage.generated_images	integer	Number of images generated
usage.output_tokens	integer	Output token count
usage.total_tokens	integer	Total token count
usage.tool_usage	object	Tool usage statistics
usage.tool_usage.web_search	integer	Web search call count (when tools_web_search is enabled)

Error Response

See Error Codes for details.

Exclusive Features of 5.0 Lite

Web-Search-Assisted Generation

Doubao Seedream 5.0 Lite is the first to support web-search-assisted image generation. When tools_web_search is enabled, the model can search internet content as a reference for image generation, significantly improving the timeliness and accuracy of generated images.

response = client.images.generate(
    model="doubao-seedream-5.0-lite",
    prompt="Latest sneaker designs of 2025",
    size="4096x4096",
    tools_web_search=True
)

3K Resolution

The exclusive 3K resolution tier strikes the best balance between image quality and generation speed, making it ideal for everyday use cases.

response = client.images.generate(
    model="doubao-seedream-5.0-lite",
    prompt="City skyline at sunset",
    size="3072x3072"
)

PNG Output

Supports setting output_format to png for scenarios that require lossless output (e.g., print materials, transparent backgrounds). Versions 4.5 and 4.0 always output jpeg.

response = client.images.generate(
    model="doubao-seedream-5.0-lite",
    prompt="Logo design, clean and modern",
    size="4096x4096",
    output_format="png"
)

Best Practices

• Web search: Ideal for time-sensitive scenarios such as trending events, latest products, and current-affairs-related image generation. Note that enabling it increases generation latency.
• 3K resolution: Best suited for everyday use, striking the optimal balance between image quality and generation speed. Switch to 4K when ultimate image fidelity is required.
• Multi-image mode: Set n > 1 or sequential_image_generation: "auto" to generate a series of images, perfect for sequential illustrations, storyboards, or stylistically consistent asset sets.
• Prompt tips: Supports mixed Chinese and English prompts. Keep prompts under 300 Chinese characters. The more specific the description, the better the output.
• Reference images: Upload reference images (up to 14) via the image parameter to guide the model toward stylistically consistent output.

Rate Limits

See Rate Limits for details.

Related Documentation

• Doubao Seedream 4.5 - Use the Seedream 4.5 version
• Doubao Seedream 4.0 - Use the Seedream 4.0 version
• Model List & Pricing - Browse available models and pricing