Q

Qdrant MCP

Vector search and semantic memory backed by Qdrant.

Qdrant·No reviews yet
Repo Docs
Share:

About

Official Qdrant MCP server that turns any Qdrant collection into a semantic memory layer for AI agents. Supports create/list/delete collections, upsert points with payload, and hybrid search with filters. Runs against self-hosted Qdrant or Qdrant Cloud.

Example tools

Illustrative tool names — the actual tool set is defined by the server at runtime.

query
Run a read-only SQL query
list_tables
List tables in the connected database
describe_table
Return column info for a table

Screenshots

No screenshots yet.

Frequently Asked Questions

Common questions about Qdrant MCP.

What is the Qdrant MCP server's primary function?
The Qdrant MCP server transforms any Qdrant collection into a semantic memory layer for AI agents. It enables vector search and semantic memory capabilities for AI applications.
How do I authenticate with the Qdrant MCP server?
Authentication for the Qdrant MCP server uses an API key. You will need to provide this key when configuring your client to connect to the server.
What operations can I perform with the Qdrant MCP server?
This MCP supports creating, listing, and deleting collections. It also allows you to upsert points with associated payloads and perform hybrid searches with filters on your Qdrant data.
Can I use the Qdrant MCP server with a self-hosted Qdrant instance?
Yes, the Qdrant MCP server is designed to run against both self-hosted Qdrant installations and Qdrant Cloud deployments. This provides flexibility in your Qdrant infrastructure.
What are the typical use cases for the Qdrant MCP server?
Typical use cases involve providing vector search and semantic memory for AI agents, enabling agents to store, retrieve, and reason over information stored in Qdrant collections.
Does using the Qdrant MCP server incur any costs?
The cost of using the Qdrant MCP server depends on your underlying Qdrant deployment. This includes whether you self-host Qdrant or utilize Qdrant Cloud, and the scale of your usage.

Install Qdrant MCP

Claude Desktop

Add this to claude_desktop_config.json.

{
  "mcpServers": {
    "qdrant": {
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-qdrant"
      ]
    }
  }
}

Cursor

Add this to ~/.cursor/mcp.json.

{
  "mcpServers": {
    "qdrant": {
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-qdrant"
      ]
    }
  }
}

VS Code

Add this to your workspace settings.json.

{
  "mcp.servers": {
    "qdrant": {
      "type": "stdio",
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-qdrant"
      ]
    }
  }
}

Reviews

Sign in to leave a review.

No reviews yet. Be the first!

Discussion(0)

Sign in to join the discussion.

No comments yet.