About volcengine
volcengine is an MCP server for using Doubao models through the Volcano Ark API. It supports text generation, multi-turn chat, embeddings, and model discovery for AI applications and semantic search.
Key features
- 01Generate text with Doubao models
- 02Hold multi-turn conversations
- 03Create text embeddings
- 04List available chat and embedding models
Use cases
- 01Generate text for Chinese-language tasks
- 02Use chat models for low-latency applications
- 03Apply reasoning models to complex analysis
- 04Build semantic search or RAG workflows
MCP support
Exposes Volcano Ark API capabilities as tools for MCP clients, including Doubao generation, chat, embeddings, and model listing. An API key is required.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
Tools
4Saved tool list · updated 3w ago
| Parameter | Type | Required | Description |
|---|---|---|---|
| model | string | No | Model ID (e.g., doubao-pro-32k, doubao-lite-4k, doubao-seed-1-6-250615) |
| prompt | string | Yes | The prompt to send to the model |
| max_tokens | number | No | Maximum tokens to generate |
| temperature | number | No | Sampling temperature (0-1) |
| system_prompt | string | No | Optional system prompt to set context |
Details
From the project READMEvolcengine-mcp
[!License: MIT](https://opensource.org/licenses/MIT) [!MCP](https://modelcontextprotocol.io) [!Node.js](https://nodejs.org)
MCP server for ByteDance Volcano Engine. Enables text generation, multi-turn chat, and embeddings via Doubao LLM models through the Volcano Ark API.
Tools (4 total)
| Tool | Description |
|---|---|
doubao_generate | Generate text using Doubao models (single-turn completion) |
doubao_chat | Multi-turn conversation with message history |
doubao_embeddings | Generate text embeddings for semantic search and RAG |
doubao_list_models | List available Doubao chat and embedding models |
Available Models
Chat Models
| Model ID | Context | Description |
|---|---|---|
doubao-pro-32k | 32K | General purpose, large context |
doubao-pro-4k | 4K | General purpose, fast |
doubao-lite-32k | 32K | Lightweight, large context |
doubao-lite-4k | 4K | Lightweight, fast |
doubao-seed-1-6-250615 | 32K | Latest with vision support |
doubao-seed-1-6-flash-250615 | 32K | Fast thinking, low latency |
doubao-seed-1-6-thinking-250615 | 32K | Deep reasoning for complex tasks |
Embedding Models
| Model ID | Dimensions | Description |
|---|---|---|
doubao-embedding | 2560 | Text embeddings |
doubao-embedding-text-240715 | 2560 | Latest text embeddings |
Install
npm install
Configuration
{
"mcpServers": {
"volcengine": {
"type": "stdio",
"command": "node",
"args": ["/path/to/volcengine-mcp/index.js"],
"env": {
"ARK_API_KEY": "your-api-key"
}
}
}
}
Environment Variables
| Variable | Description | Required |
|---|---|---|
ARK_API_KEY | Volcano Ark API key | Yes |
VOLCENGINE_API_KEY | Alternative API key variable | Yes (if ARK_API_KEY not set) |
ARK_BASE_URL | API base URL | No (defaults to Beijing endpoint) |
ARK_CHAT_MODEL | Default chat model | No (defaults to doubao-pro-32k) |
ARK_EMBEDDING_MODEL | Default embedding model | No (defaults to doubao-embedding) |
Getting an API Key
- Sign up at console.volcengine.com
- Complete real-name authentication (required for Chinese cloud services)
- Navigate to Large Models > Volcano Ark > API Key Management
- Create and copy your API key
Use Cases
- Chinese language tasks -- Doubao excels at Chinese NLP
- Low-latency inference -- Doubao Flash for real-time applications
- Deep reasoning -- Doubao Thinking for complex analysis
- Cost-effective bulk processing -- Doubao Lite for high-volume tasks
- Semantic search and RAG -- Doubao embeddings (2560 dimensions)
Dependencies
@modelcontextprotocol/sdk-- MCP protocol SDK
License
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.