Seedream 4.5 Text-to-Image - Starrise AI

Seedream 4.5 Text-to-Image is a text-to-image model from ByteDance’s Seedream series, available through the Starrise AI API.

Key Capabilities

Text-to-image — Generate images from text prompts
Custom size — Flexible output dimensions
Watermark control — Optional watermark and logo configuration

Quick Example

curl https://ai.alad.com/v1/images/generations \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "doubao-seedream-4-5-251128",
    "prompt": "A serene mountain landscape at sunrise, mist filling the valley",
    "size": "1024x1024",
    "watermark": false
  }'

import requests

response = requests.post(
    "https://ai.alad.com/v1/images/generations",
    headers={
        "Authorization": "Bearer YOUR_API_KEY",
        "Content-Type": "application/json"
    },
    json={
        "model": "doubao-seedream-4-5-251128",
        "prompt": "A serene mountain landscape at sunrise, mist filling the valley",
        "size": "1024x1024",
        "watermark": False
    }
)

task = response.json()
print(f"Task ID: {task['id']}")

Parameters

Parameter	Type	Required	Description
`model`	string	Yes	Must be `doubao-seedream-4-5-251128`
`prompt`	string	Yes	Image generation prompt
`size`	string	No	Output image size, e.g. `1024x1024`, `2K`
`watermark`	boolean	No	Default `false`
`seed`	integer	No	Random seed
`logo_info`	object	No	Logo config: `add_logo`, `position`, `language`, `opacity`

API Reference

View the interactive API Playground for Seedream 4.5 Text-to-Image.