Qdrant MCP Server

The official Qdrant MCP Server provides persistent semantic memory and high-performance vector search to AI assistants.

Capabilities

• Memory Storage: Store text snippets and metadata payloads with automatic embeddings via FastEmbed (all-MiniLM-L6-v2).
• Semantic Retrieval: Execute similarity searches across vector collections to retrieve relevant past context.
• Multi-Collection Management: Partition memory spaces across separate collections.
• Flexible Transport: Supports standard input/output (stdio), sse, and streamable-http transport modes.

Configuration

Configure your client to point to a local or cloud Qdrant cluster:

{
  "mcpServers": {
    "qdrant": {
      "command": "uvx",
      "args": ["mcp-server-qdrant"],
      "env": {
        "QDRANT_URL": "http://localhost:6333",
        "COLLECTION_NAME": "agent-memories"
      }
    }
  }
}