Qdrant Vector Database

v1.0.0Databasesstable

Enables semantic search and document management using a local Qdrant vector database with OpenAI embeddings. Supports natural language queries, metadata filtering, and collection management for AI-powered document retrieval.

claudeembeddingsmcpmodel-context-protocolopenai
Share:
32
Stars
0
Downloads
0
Weekly
0/5

What is Qdrant Vector Database?

Qdrant Vector Database is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to enables semantic search and document management using a local qdrant vector database with openai embeddings. supports natural language queries, metadata filtering, and collection management for ai-pow...

Enables semantic search and document management using a local Qdrant vector database with OpenAI embeddings. Supports natural language queries, metadata filtering, and collection management for AI-powered document retrieval.

This server falls under the Databases category on MCPgee, the world's largest MCP server directory with 33,000+ servers.

Features

  • Enables semantic search and document management using a loca

Use Cases

Semantic search
Document management
AI-powered retrieval
mhalder

Maintainer

LicenseMIT License
Languagetypescript
Versionv1.0.0
UpdatedMay 16, 2026
Statushealthy
Maintenanceactive

Works with

ClaudeOpenAIwindowsmacoslinux

Installation

Manual Installation

npx qdrant-mcp-server

Configuration

Configuration Details

Config File

claude_desktop_config.json

Performance

Response Metrics

Response Time< 200ms
ThroughputMedium

Resource Usage

Memory UsageLow
CPU UsageLow

How to Set Up and Use Qdrant Vector Database

The Qdrant MCP Server connects AI assistants to a local Qdrant vector database, enabling semantic and hybrid search over documents and codebases using embedding providers like OpenAI, Cohere, Voyage, or a local Ollama model. It exposes tools for collection management, document ingestion, AST-aware code vectorization across 35+ programming languages, and Git history indexing. Developers use it to give their AI assistant a persistent, searchable memory of their codebase and project documents without relying on external cloud search services.

Prerequisites

  • Node.js 18+ (Node 24.x requires CXXFLAGS='-std=c++20' during install)
  • Docker or Podman to run the Qdrant vector database container
  • An embedding provider: local Ollama with nomic-embed-text pulled, or an API key for OpenAI, Cohere, or Voyage
  • Git (optional, for Git history indexing features)
  • An MCP-compatible client such as Claude Desktop or the Claude CLI
1

Clone and install dependencies

Clone the repository and install Node.js dependencies. On Node 24.x you must set the C++ standard flag before installing.

git clone https://github.com/mhalder/qdrant-mcp-server.git
cd qdrant-mcp-server
npm install
# Node 24.x only:
# CXXFLAGS='-std=c++20' npm install
2

Start the Qdrant vector database

Use the included Docker Compose file to launch a local Qdrant instance. If using Ollama for embeddings, also pull the embedding model.

docker compose up -d
# For Ollama embeddings:
docker exec ollama ollama pull nomic-embed-text
3

Build the server

Compile the TypeScript source into the build directory before running.

npm run build
4

Configure environment variables

Set the embedding provider and any required API keys. QDRANT_URL defaults to http://localhost:6333. EMBEDDING_PROVIDER can be ollama, openai, cohere, or voyage.

export EMBEDDING_PROVIDER=openai
export OPENAI_API_KEY=sk-...
export QDRANT_URL=http://localhost:6333
# Optional:
export LOG_LEVEL=info
5

Register the server with your MCP client

Add the server to your Claude Desktop config file or use the Claude CLI to register it over stdio.

claude mcp add --transport stdio qdrant -- node /path/to/qdrant-mcp-server/build/index.js
6

Index your codebase

Once connected, ask your AI assistant to index a project directory. The server will chunk and embed your code using AST-aware splitting.

Qdrant Vector Database Examples

Client configuration

Claude Desktop config using the stdio transport with OpenAI embeddings.

{
  "mcpServers": {
    "qdrant": {
      "command": "node",
      "args": ["/path/to/qdrant-mcp-server/build/index.js"],
      "env": {
        "EMBEDDING_PROVIDER": "openai",
        "OPENAI_API_KEY": "sk-...",
        "QDRANT_URL": "http://localhost:6333"
      }
    }
  }
}

Prompts to try

Example prompts once the server is connected and a collection or codebase has been indexed.

- "Index the codebase at /home/user/myproject and create a collection called myproject"
- "Search the myproject collection for code related to authentication middleware"
- "Find all usages of the UserService class across the indexed codebase"
- "Search the Git history for commits that modified the payment module"
- "List all available collections in my Qdrant instance"

Troubleshooting Qdrant Vector Database

npm install fails on Node 24.x with a C++ compilation error

Set the C++ standard flag before installing: CXXFLAGS='-std=c++20' npm install

Embedding calls fail with an authentication error

Ensure the correct API key environment variable is set for your chosen provider: OPENAI_API_KEY, COHERE_API_KEY, or VOYAGE_API_KEY. Confirm EMBEDDING_PROVIDER matches the key you provided.

Qdrant connection refused at http://localhost:6333

Run 'docker compose up -d' from the repository root to start the Qdrant container, then verify it is healthy with 'docker ps'.

Frequently Asked Questions about Qdrant Vector Database

What is Qdrant Vector Database?

Qdrant Vector Database is a Model Context Protocol (MCP) server that enables semantic search and document management using a local qdrant vector database with openai embeddings. supports natural language queries, metadata filtering, and collection management for ai-powered document retrieval. It connects AI assistants to external tools and data sources through a standardized interface.

How do I install Qdrant Vector Database?

Follow the installation instructions on the Qdrant Vector Database GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.

Which AI clients work with Qdrant Vector Database?

Qdrant Vector Database works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.

Is Qdrant Vector Database free to use?

Yes, Qdrant Vector Database is open source and available under the MIT License license. You can use it freely in both personal and commercial projects.

Browse More Databases MCP Servers

Explore all databases servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.

Quick Config Preview

{ "mcpServers": { "qdrant-mcp-server": { "command": "npx", "args": ["-y", "qdrant-mcp-server"] } } }

Add this to your claude_desktop_config.json or .cursor/mcp.json

Read the full setup guide →

Ready to use Qdrant Vector Database?

Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.

33,000+ ServersFree & Open SourceStep-by-Step Guides