mcp-server-qdrant

mcp-server-qdrant

The Qdrant Model Context Protocol (MCP) server implementation for storing and retrieving information in a vector database.

MCPDataOpen source
Type
MCP
Transport
http
Open source
Yes
GitHub Stars
★ 1.4k
Source
mcp-github

Overview

mcp-server-qdrant is the official MCP server implementation for the Qdrant vector search engine. It acts as a semantic memory layer on top of the Qdrant database, enabling AI applications to seamlessly integrate and access external data sources. With this capability, AI can store and retrieve relevant information, making it suitable for use cases requiring efficient vector search. Configuration is done via environment variables such as Qdrant server URL and API Key, and the server runs using a specified transport protocol. Ideal for developers needing efficient vector search and semantic memory functionality.

Capabilities

  • Store information in Qdrant
  • Retrieve relevant information from Qdrant
  • Support custom collection names
  • Support multiple embedding models

Use cases

AI-powered chatbotsIntelligent recommendation systemsSemantic search applicationsKnowledge graph construction

Setup

Requires: API KeyNode 环境
Configure the Qdrant server URL and API Key via environment variables, then run mcp-server-qdrant using `uvx`.

This information was compiled by AI from public sources and may contain inaccuracies — please refer to the source.

FAQ

How do I configure the Qdrant server?

By setting the environment variables `QDRANT_URL` and `QDRANT_API_KEY`.

Which embedding models are supported?

Currently supports `fastembed` and `sentence-transformers/all-MiniLM-L6-v2`.

Related skills