home / mcp / doubao image mcp server
An MCP server that generates images using the Doubao model via Volcano Engine, with local saving, logging, and MCP-compatible tooling.
Configuration
View docs{
"mcpServers": {
"suibin521-doubao-image-mcp-server": {
"url": "https://ark.cn-beijing.volces.com/api/v3",
"headers": {
"BASE_URL": "https://ark.cn-beijing.volces.com/api/v3",
"API_MODEL_ID": "ep-20250528154802-c4np4",
"DOUBAO_API_KEY": "YOUR_API_KEY",
"IMAGE_SAVE_DIR": "C:/images"
}
}
}
}You run a dedicated MCP server that interfaces with the Doubao image generation model to produce high-quality images. It supports multiple resolutions, bilingual prompts, local image saving, and detailed logging, making it easy to integrate image generation into your AI assistant workflows.
You interact with the Doubao image MCP server through an MCP client. After you configure the MCP connection, simply describe the image you want and the server will generate it using the Doubao model. The server saves generated images to your specified directory and returns the results to your client. You can customize prompts, resolution, seed, guidance scale, and watermark settings for precise results.
Prerequisites you need before installation: Python 3.13 or newer, access to Volcano Engine API, and an active inference endpoint ID.
Step 1: Install the MCP server dependencies locally. You can install via a package manager or a development tool, depending on your preference.
Step 2: Prepare the MCP configuration file that points to your server and environment variables. The server expects configurations through the MCP config, and environment values are passed via the env field.
Step 3: Start the server. You can run the server directly in development with Python or use an MCP runner tool to host it as an MCP service.
Environment variables you need to set in your MCP config under the env field are the following: BASE_URL, DOUBAO_API_KEY, API_MODEL_ID, and IMAGE_SAVE_DIR. BASE_URL is the Volcano Engine API base address, DOUBAO_API_KEY is your API key from Volcano Engine, API_MODEL_ID identifies your inference endpoint, and IMAGE_SAVE_DIR is the local folder where images will be saved. Ensure IMAGE_SAVE_DIR exists or the server will create it automatically.
When you start the server, you can also configure multiple image generation parameters such as size, seed, guidance_scale, and watermark. The server supports resolutions from 512x512 up to 2048x2048 and can generate images in formats like JPG.
1) Configure MCP with BASE_URL, DOUBAO_API_KEY, API_MODEL_ID, and IMAGE_SAVE_DIR.
2) Start the server using your preferred method (see installation steps). Then connect your MCP client and issue a generate image request with your desired prompt and options.
Main tool for generating images. Accepts prompt, size, seed, guidance_scale, watermark, and file_prefix to produce and save images.