qdrant-mcp-server

qdrant-mcp-server

0
databases

The Qdrant MCP Server is designed to interface with the Qdrant vector database, offering tools for managing and searching vector data. Key features include automatic text-to-vector embedding and efficient text similarity searches.

Qdrant MCP Server

An MCP server for the Qdrant vector database that facilitates vector management and similarity searches using automatic text-to-vector embedding.

Features

  • Automatic text-to-vector embedding using FastEmbed
  • Text content storage and retrieval with vector search
  • Default collection configuration through environment variables
  • Text similarity search by content
  • Efficient embedding with optimized models

Tools

Text Tools

  • store_text: Store text as an embedding vector
  • search_similar_text: Find similar vectors for a given query
  • store_texts: Batch store multiple texts with embeddings

Vector Tools

  • search_vectors: Search for similar vectors
  • upsert_vectors: Upload vectors to a collection
  • filter_search: Apply metadata filters during search

Point Tools

  • get_points: Retrieve points by IDs
  • delete_points: Remove points by IDs
  • count_points: Count points in a collection

Usage

Running locally

  1. Install the package
  2. Run the server

Running with Docker

  1. Build the Docker image
  2. Run the container

Testing

  • Use provided test suite to validate functionality