OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Agent Memory & RAG

Qdrant MCP on Raspberry Pi 5

Qdrant MCP on Raspberry Pi 5 is a JavaScript developer integration that connects MCP clients to Qdrant vector storage for persistent, semantic memory. Its documented local mode stores data on the Raspberry Pi using Qdrant’s local SQLite-backed storage, while an alternative configuration connects to a containerized Qdrant server.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Qdrant MCP on Raspberry Pi 5 — OpenViST project profile in Agent Memory & RAG.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when an agent or other MCP client needs a persistent semantic memory store, especially when you want the documented local-storage setup on Raspberry Pi 5 or need the separately described containerized Qdrant option.

Who it is for

  • People configuring persistent memory for MCP-compatible AI agents.
  • Raspberry Pi 5 users who want locally hosted semantic storage and search.
  • OpenClaw or Claude Code users exploring agent memory integrations.
  • Individuals or teams organizing personal notes, shared information, or project context for semantic retrieval.

Documented uses

  • Store facts, decisions, corrections, or notes so they can be retrieved in later interactions.
  • Search saved information by meaning when a query uses different wording from the stored text.
  • Bulk-load memories with the documented seed script or separate work and personal information into named collections.
  • Connect an agent to semantic memory through an MCP client or, for the documented OpenClaw plugin, inject retrieved memories before response generation.
Before you choose

Requirements & limits

  • The local setup requires an absolute QDRANT_LOCAL_PATH and a collection name.
  • mcporter is the documented MCP client or bridge; the README also lists the Qdrant MCP server installation through pip.
  • Local embedded storage and a remote Qdrant server are configured differently and cannot use QDRANT_LOCAL_PATH and QDRANT_URL simultaneously.
  • The README contains differing embedding details: its stack section describes Gemini gemini-embedding-001 via API, while its how-it-works and configuration sections describe all-MiniLM-L6-v2. Confirm the configuration you
  • The project lists OpenClaw as optional; its README notes an OpenClaw version-specific limitation around direct mcpServers configuration and describes mcporter as the bridge.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Configure local Qdrant memory with mcporter

Set up the documented local Qdrant MCP server configuration and use its memory store and search operations.

Before you start

  • Raspberry Pi 5; the README lists an 8 GB RAM configuration.
  • Python pip and Node.js npm for installing the Qdrant MCP server and mcporter.
  • An absolute local storage path and a chosen Qdrant collection name.

Steps

  1. Install the Qdrant MCP server with the command below and mcporter with the command below.

    Documented command
    pip3 install mcp-server-qdrant
    Documented command
    npm install -g mcporter

  2. Create the mcporter configuration directory with mkdir -p ~/.mcporter.

  3. Create ~/.mcporter/mcporter.json with an MCP server entry that runs mcp-server-qdrant and sets QDRANT_LOCAL_PATH to an absolute path and COLLECTION_NAME to the desired collection.

  4. Use mcporter list to check that the server is registered, then call qdrant-store to save information and qdrant-find to search it.

Expected outcome

An mcporter-configured local Qdrant MCP server with documented calls for storing and semantically searching agent memories.

Guide limitations

  • The guide documents local mode; it does not verify a completed installation or successful calls.
  • Use an absolute value for QDRANT_LOCAL_PATH; the README warns that tilde expansion does not work in mcporter environment variables.
  • If mcporter cannot find the server, the README suggests trying the full path to the server command.
  • Do not set QDRANT_URL alongside QDRANT_LOCAL_PATH; the README treats server mode and embedded local mode as alternatives.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
15 October 2026
Original website title
rockywuest/qdrant-mcp-pi5 · Official homepage title · retained in its source language
How we review sourcesSuggest a correction