MCP Memory Server: What It Is & How to Choose (2026)

Anthony Conti of Astra AI, LLC published a guide to MCP memory servers, explaining their role in the Model Context Protocol ecosystem and how they differ from vector databases and RAG. The article compares various server options and outlines key selection criteria, including the author's own MIND server.

Originally published on m-i-n-d.ai https://www.m-i-n-d.ai/mcp-memory-server . I write and maintain MIND, an MCP memory server — this piece tries to be straight about where MIND fits and where it doesn't. The definition, the architecture, how to choose one, and an honest comparison of every option — including the GitHub repos that outrank us on this term today. By Anthony Conti · Astra AI, LLC · Last updated September 7, 2026 An MCP memory server is a Model Context Protocol https://modelcontextprotocol.io/introduction server whose job is to store and retrieve memories on behalf of an AI agent — facts, preferences, decisions, past conversations — so that what an agent learns in one session, on one tool, is still there the next time it is asked, on whatever tool asks it. MCP is the open standard Anthropic open-sourced on November 25, 2024 for connecting AI applications to external tools and data, and it is now supported by Claude, ChatGPT, Cursor, VS Code, Windsurf and dozens of other clients. A memory server is one specific kind of MCP server: instead of exposing a calendar or a Git repository, it exposes memory itself, as two or three callable tools — typically something shaped like store memory and search memory . The one-sentence version. Without MCP, every AI tool you use has to reinvent memory, and none of them can read what another one remembers. With an MCP memory server in the middle, memory becomes a service any MCP-compatible agent can call — the same way any browser can render any website because HTTP is a shared standard. Your AI Agent ←→ MCP JSON-RPC over HTTP or stdio ←→ MCP Memory Server ←→ Storage Claude, Cursor, two or three tools: this is the part vector DB, Windsurf, ChatGPT store memory / search memory that varies graph, files... No — this is the single most common confusion, so it is worth being precise about. A vector database Pinecone, Weaviate, Qdrant, pgvector is a storage engine: you send it embeddings, it returns nearest neighbours. It has no concept of a tool call, a session, or an agent. An MCP memory server is the layer above that: it decides what gets stored, when, and how it is exposed to an agent as a callable tool. Several servers in the comparison table below use a vector database as their storage engine internally Puliczek/mcp-memory runs on Cloudflare Vectorize, for instance — the vector database itself has no idea MCP exists. They overlap but answer different questions. Retrieval-augmented generation is a technique: fetch the most relevant chunks of text and stuff them into the prompt before generating an answer. An MCP memory server is an integration surface: a standard interface an agent calls as a tool, regardless of what retrieval technique runs behind it. Most MCP memory servers use RAG-style retrieval internally, but RAG is the algorithm and MCP is the protocol carrying the request. You can build a memory server with no RAG at all a flat key-value store would still qualify , and you can do RAG with no MCP in sight most chatbot RAG pipelines predate MCP entirely . Every MCP memory server, however it stores data underneath, is built from the same three MCP primitives, formalised in the 2025-06-18 specification https://modelcontextprotocol.io/specification/2025-06-18/server/tools and refined again in the 2025-11-25 revision: store memory , search memory , sometimes delete memory or list memories . Since the 2025-06-18 spec, tools can declare a structured output schema, so the agent gets typed results back instead of a raw string. stdio the server runs as a child process on your machine — simplest, single-user or remote HTTP the server runs somewhere else and the agent authenticates over OAuth — multi-device, multi-tool . Five questions narrow the field fast. Most projects fail at least one of them for most people, which is fine — they were not built to pass all five. Six real, currently-maintained options, verified against their own repos and pricing pages this session — MIND included, and not flattered. Every one of these is legitimately good at something. | Project | What it stores memory in | Hosting model | Worth knowing | |---|---|---|---| | MIND https://www.m-i-n-d.ai/mcp.html | Knowledge graph entities + relationships + embeddings , MongoDB-backed | Remote one-paste OAuth or self-hosted local via npm — both free tiers | Memory plus a real product web/mobile app on top; 50+ LLM models can read the same graph | | Mem0 via coleam00/mcp-mem0 https://github.com/coleam00/mcp-mem0 | Mem0's own vector + graph memory store | Template server you run yourself, pointed at Mem0 Cloud or self-hosted Mem0 | A reference/template implementation, not an official Mem0 product — good starting point to fork | | okooo5km/memory-mcp-server https://github.com/okooo5km/memory-mcp-server | A local knowledge graph entities, relations, observations in a JSON file | Local only, runs as a stdio MCP server | Closest thing to a reference implementation for graph-shaped memory over MCP; also ships a Go port | | JamesANZ/memory-mcp https://github.com/JamesANZ/memory-mcp | MongoDB-backed memories with tag-based search | Self-hosted; you provide the MongoDB connection | Built for coding-agent context caching across Cursor and Claude Desktop specifically | | Puliczek/mcp-memory https://github.com/Puliczek/mcp-memory | Vector embeddings on Cloudflare Vectorize, metadata in D1 | Cloudflare Workers — you deploy your own instance | Purpose-built for remembering user preferences/behaviors, not general knowledge capture | | Zep Memory MCP Server seats https://www.getzep.com/pricing | Temporal knowledge graph via Zep's Graphiti engine | Cloud only — Zep retired its self-hosted Community Edition Graphiti itself stays open source | Bills by 'credits' roughly 1 credit per 350 bytes ingested ; Free tier caps at 1 MCP seat | | Letta formerly MemGPT https://github.com/letta-ai/letta | An OS-style memory model — active context is 'RAM', archival memory is 'disk' | Fully open source and self-hostable, or Letta Cloud | 23,081 GitHub stars as of this writing — the most-starred project in this table by a wide margin | Star counts and pricing were checked against each project's own GitHub repo or pricing page during this session 2026-09-07 ; all three change constantly — verify against the source before deciding. Almost every project in the table above converges on the same small tool surface, even though the storage underneath is completely different from one to the next: Worth knowing if you are betting infrastructure on this protocol: Anthropic open-sourced MCP as a company project in November 2024, but has since moved to hand off governance. Anthropic announced it is donating MCP to a newly established Agentic AI Foundation, moving the specification out from under a single vendor and toward the kind of neutral stewardship that standards like HTTP or OAuth have. For anyone choosing an MCP memory server today, that is a modest but real signal that the protocol itself is built to outlast any one company's roadmap — including Astra AI's. Full disclosure: this page is published by Astra AI, LLC, which builds MIND. Here is the honest version anyway. Every project above except MIND stops at the server — there is infrastructure, and then it's your job to build something a person actually opens. MIND is an MCP memory server connect it in one paste at /mcp-memory-server/setup https://www.m-i-n-d.ai/mcp-memory-server/setup backed by a real knowledge graph, with a web and mobile app on top, so the same memory Claude Code just wrote is visible to you, searchable, and readable by 50+ other LLM models the next time you switch tools. If you specifically want the leanest possible memory API for one coding agent and nothing else, Mem0 or Letta may be the better fit — we would rather say that than sell you the wrong tool. Is an MCP memory server the same as a vector database? No. A vector database Pinecone, Weaviate, Qdrant, pgvector is a storage engine for embeddings — it answers similarity queries. An MCP memory server sits on top of some storage engine a vector database, a knowledge graph, a plain document store, or all three and exposes it to an AI agent through the Model Context Protocol's tool-calling interface. Several of the servers in the table above use a vector database internally; the vector database itself has no idea what MCP is. Is an MCP memory server the same as RAG? They overlap but answer different questions. Retrieval-augmented generation is a technique — fetch relevant chunks, stuff them into the prompt. An MCP memory server is an integration surface — a standard way for an agent to call store memory and search memory as tools, regardless of what retrieval technique runs underneath. Most MCP memory servers use RAG-style retrieval internally, but MCP is the protocol, not the algorithm. Can I just use a local JSON file instead of a real memory server? For a single-machine, single-user coding assistant that never needs to search or grow past a few hundred entries, yes — several of the projects above started exactly there. It stops working the moment you need semantic search over the memories, multiple agents writing concurrently, or memory that follows you to a second machine. Does MIND compete with Mem0, Zep, and Letta? Partially, and we say so on purpose. Mem0, Zep, and Letta are excellent developer-infrastructure choices if you are building a single coding agent and want the leanest possible memory API. MIND is the pick if a person, not just one agent, needs to open a product, see their own graph, and have every AI they use — not one integrated app — read and write to it. Sources verified this session: Anthropic — Introducing the Model Context Protocol https://www.anthropic.com/news/model-context-protocol , the MCP 2025-06-18 tools specification https://modelcontextprotocol.io/specification/2025-06-18/server/tools , Anthropic — donating MCP to the Agentic AI Foundation https://www.anthropic.com/news/donating-the-model-context-protocol-and-establishing-of-the-agentic-ai-foundation , and each listed project's own repository or pricing page. Full piece, kept live and updated, is here: m-i-n-d.ai/mcp-memory-server https://www.m-i-n-d.ai/mcp-memory-server . If you want the 10-minute MCP setup guide or the docs for MIND itself, that's at m-i-n-d.ai/mcp.html https://www.m-i-n-d.ai/mcp.html .