Seedream MCP Integration Guide

MCP (Model Context Protocol) is a model context protocol launched by Anthropic that allows AI models (such as Claude, GPT, etc.) to call external tools through a standardized interface. With the Seedream MCP Server provided by 费思量-API, you can directly use ByteDance Seedream to generate and edit AI images in AI clients like Claude Desktop, VS Code, Cursor, and more.

Feature Overview

Seedream MCP Server offers the following core features:

  • Text-to-Image Generation — Generate high-quality images from text prompts (supports both Chinese and English prompts)
  • Image Editing — Modify existing images’ style, background, attributes, etc.
  • Virtual Try-On — Virtually wear clothes on person photos
  • Multi-Model Support — Supports Seedream v4.5, v4.0, v3.0 T2I, SeedEdit v3.0 I2I, and other models
  • Multiple Resolutions — Supports 1K, 2K, 4K, and adaptive resolutions
  • Seed Control — Achieve reproducible generation results via the seed parameter
  • Task Query — Monitor generation progress and retrieve results

Prerequisites

Before use, you need to obtain an 费思量-API API Token:

  1. Register or log in to the 费思量-API Platform
  2. Go to the Seedream Images API page
  3. Click "Acquire" to get your API Token (first-time applicants receive free quota)

Installation and Configuration

Method 1: pip Installation (Recommended)

pip install mcp-seedream-pro

Method 2: Source Installation

git clone https://github.com/AceDataCloud/SeedreamMCP.git
cd SeedreamMCP
pip install -e .

After installation, you can start the service using the mcp-seedream-pro command.

Usage in Claude Desktop

Edit the Claude Desktop configuration file:

  • macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
  • Windows: %APPDATA%\Claude\claude_desktop_config.json

Add the following configuration:

{
  "mcpServers": {
    "seedream": {
      "command": "mcp-seedream-pro",
      "env": {
        "ACEDATACLOUD_API_TOKEN": "your API Token"
      }
    }
  }
}

If using uvx (no need to pre-install packages):

{
  "mcpServers": {
    "seedream": {
      "command": "uvx",
      "args": ["mcp-seedream-pro"],
      "env": {
        "ACEDATACLOUD_API_TOKEN": "your API Token"
      }
    }
  }
}

Save the configuration and restart Claude Desktop to use Seedream-related tools in conversations.

Usage in VS Code / Cursor

Create .vscode/mcp.json in the project root:

{
  "servers": {
    "seedream": {
      "command": "mcp-seedream-pro",
      "env": {
        "ACEDATACLOUD_API_TOKEN": "your API Token"
      }
    }
  }
}

Or use uvx:

{
  "servers": {
    "seedream": {
      "command": "uvx",
      "args": ["mcp-seedream-pro"],
      "env": {
        "ACEDATACLOUD_API_TOKEN": "your API Token"
      }
    }
  }
}

Available Tools

Tool Name Description
seedream_generate_image Generate images from text prompts
seedream_edit_image Edit or modify existing images
seedream_get_task Query the status of a single task
seedream_get_tasks_batch Batch query task statuses
seedream_list_models List all available models and their capabilities
seedream_list_sizes List available image sizes and resolution options

Usage Examples

After configuration, you can directly invoke these features in AI clients using natural language, for example:

  • "Help me generate a landscape painting in ink wash style using Seedream"
  • "Change the background of this photo to a starry sky"
  • "Generate a 4K high-definition portrait using Seedream v4.5"
  • "Virtually try this clothing on the person in this photo"

More Information