Better Qdrant MCP Server logo

Better Qdrant MCP Server

by wrediam

The Better Qdrant MCP Server enhances Qdrant vector database functionality by providing tools for managing collections, adding documents, and performing semantic searches. It supports various embedding services like OpenAI, OpenRouter, Ollama, and FastEmbed.

View on GitHub

Last updated: N/A

What is Better Qdrant MCP Server?

A Model Context Protocol (MCP) server designed to extend and improve the capabilities of the Qdrant vector database. It allows users to manage Qdrant collections, add documents for indexing, and execute semantic searches using different embedding services.

How to use Better Qdrant MCP Server?

  1. Install the server using npm install -g better-qdrant-mcp-server or npx better-qdrant-mcp-server. 2. Configure environment variables in a .env file (QDRANT_URL, QDRANT_API_KEY, and API keys for your chosen embedding services). 3. Integrate the server into your application or use it with Claude as shown in the example configuration. 4. Use the provided example commands to list collections, add documents, search, or delete collections.

Key features of Better Qdrant MCP Server

  • List Collections

  • Add Documents with multiple embedding services

  • Perform Semantic Searches

  • Delete Collections

Use cases of Better Qdrant MCP Server

  • Semantic search applications

  • Document indexing and retrieval systems

  • AI-powered chatbots and assistants

  • Knowledge base management

FAQ from Better Qdrant MCP Server

What is the default embedding service?

The default embedding service is not explicitly defined in the README, but the example configuration suggests Ollama is a common choice.

What are the requirements to run this server?

Node.js >= 18.0.0, a running Qdrant server, and API keys for the embedding services you want to use.

How do I configure the server?

The server is configured using environment variables, which can be set in a .env file.

Which embedding services are supported?

OpenAI, OpenRouter, Ollama, and FastEmbed are supported.

How do I add documents to a collection?

Use the add_documents tool with the required arguments, including the file path, collection name, embedding service, chunk size, and chunk overlap.