Back to MCP Servers
🔌

elvismdev/mem0-mcp-selfhosted

MITAPI key required

](https://glama.ai/mcp/servers/elvismdev/mem0-mcp-selfhosted) 🐍 🏠 🍎 🪟 🐧 - Self-hosted mem0 MCP server for Claude Code with Qdrant vector search, Neo4j knowledge graph, and Ollama embeddings. Zero

PythonSearch

Install

uvx --from

Required environment variables

MEM0_ANTHROPIC_TOKENSet in your MCP config
ANTHROPIC_API_KEYSet in your MCP config
MEM0_LLM_MAX_TOKENSet in your MCP config
GOOGLE_API_KEYSet in your MCP config
MEM0_QDRANT_API_KEYSet in your MCP config

claude_desktop_config.json

claude_desktop_config.json
{
  "mcpServers": {
    "elvismdev-mem0-mcp-selfhosted": {
      "command": "uvx",
      "args": [
        "--from"
      ],
      "env": {
        "MEM0_ANTHROPIC_TOKEN": "<YOUR_MEM0_ANTHROPIC_TOKEN>",
        "ANTHROPIC_API_KEY": "<YOUR_ANTHROPIC_API_KEY>",
        "MEM0_LLM_MAX_TOKEN": "<YOUR_MEM0_LLM_MAX_TOKEN>",
        "GOOGLE_API_KEY": "<YOUR_GOOGLE_API_KEY>",
        "MEM0_QDRANT_API_KEY": "<YOUR_MEM0_QDRANT_API_KEY>"
      }
    }
  }
}

Add this to your Claude Desktop config file. Find it at ~/Library/Application Support/Claude/claude_desktop_config.json on macOS.

What this does

elvismdev/mem0-mcp-selfhosted exposes a set of tools to Claude over the Model Context Protocol. After you add it to claude_desktop_config.json (snippet above) and restart Claude Desktop, those tools become callable inside any conversation. That makes it useful when you want Claude to reach into search engines and knowledge bases without copy-pasting context every turn.

Requirements

This server needs the following environment variables to be set before it can run:

MEM0_ANTHROPIC_TOKEN
ANTHROPIC_API_KEY
MEM0_LLM_MAX_TOKEN
GOOGLE_API_KEY
MEM0_QDRANT_API_KEY

Set these via the env object in your MCP config (see claude_desktop_config.json snippet above).

Common use cases

  • Search the open web, internal docs, or a vector store from inside a Claude session
  • Bring fresh information past Claude's training cutoff into context
  • Run multi-source research without copy-pasting URLs and snippets