Cognee vs Memlo: which memory to use

Last updated October 31, 2026

Cognee is an open-source AI memory engine: it turns documents, code and conversations into a knowledge graph plus vector search that agents query, and you can run it yourself or use Cognee's cloud. Memlo is a hosted memory server for the agents you already use: Claude Code, Codex, claude.ai, ChatGPT and other MCP apps share memories that agents save on purpose, each with its own version history. Pick Cognee to build graph memory over your own data, or to self-host. Pick Memlo to give your own agents and your team one shared memory with nothing to run.

At a glance

Facts about Cognee come from its own repository, documentation and pricing page, read on October 31, 2026. Links are in Sources at the end.

How Cognee works

Cognee's README says it "gives AI agents persistent long-term memory across sessions," turning "documents, code, and conversations into a self-hosted knowledge graph." Its main operations are remember, recall, improve and forget; at query time, "retrieval selects relevant graph, vector, or code context." Older guides use add, cognify and search, which Cognee's documentation now lists as legacy operations.

Run locally, Cognee uses a file-based graph database (Ladybug, which also answers to kuzu) and LanceDB for vectors by default. Neo4j is the recommended graph store for production, and Postgres, Amazon Neptune, PGVector, ChromaDB and others are supported. With Docker, the API, a web UI and the MCP server run on ports 8000, 3000 and 8001.

The cognee-mcp server exposes four tools: remember, recall, forget and cognify_status. It runs over stdio by default, or over SSE or Streamable HTTP.

Cognee Cloud has a Free plan (one workspace and 1M tokens included, unlimited users) and a Standard plan at $1.00 per 1M tokens plus $5 per additional workspace. Enterprise is "delivered as a BYOC engagement," with no listed price.

How Memlo works

Memlo is an MCP server at https://memlo.ai/mcp. You add it to each agent and sign in once, with no API key. Agents get tools to save, search, read, change and retire memories.

• Curated memories. Each memory is a short note with a name and a title, saved when the agent learns something worth keeping.
• History, with authors. Every change is a version, with when it was written and by whom. Any version can be restored, and edit_memory refuses a write when someone else changed the note first.
• Spaces and sharing. Keep memories in spaces, and share a space or a page with people by email as a Viewer or an Editor.
• A wiki and a graph. Memlo keeps a page for each person, project and tool your memories mention, with every line citing its memory.

Where Cognee is the better choice

• You're building on your own data. Cognee is made to turn documents, code and conversations into a graph your agent queries.
• You need to self-host. The full engine is open source and runs locally or on your own servers. Memlo is hosted only.
• You want a library or API. Cognee is a Python library with an API; Memlo has no SDK.
• You want to change the code. Cognee is Apache 2.0.
• You want to start today. It's open source with a free cloud plan; Memlo is by invitation.

Where Memlo is the better choice

• You want memory for the agents you already use, with no setup. Add one URL to Claude Code, Codex, claude.ai or ChatGPT and sign in. There's no database or service to run.
• You want to read and correct each memory. Memories are short titled notes you can open, edit and roll back, not chunks and graph nodes produced by a pipeline.
• You want to see which agent wrote what, and roll it back. Every version names the person and the connected app or agent, and any version can be restored.
• You share with people by role. Share a space or a single page by email as a Viewer or an Editor.
• You want hosted sign-in for every app. Each app signs in with OAuth and can be disconnected on its own.

What Memlo doesn't do yet

• Cursor. Memlo's sign-in doesn't work from Cursor yet.
• Automatic capture. Agents save memories on purpose; nothing processes everything you add into a graph in the background.
• An SDK or API keys. Memlo is reached over MCP with your own sign-in, while Cognee is a library with an API.
• Self-hosting. Memlo is hosted only.
• Open sign-up. Accounts are by invitation for now; you can join the waitlist.

Frequently asked questions

Is Cognee open source?

Yes. Cognee is under the Apache 2.0 license. Its pricing page says you can "run the full memory engine locally… free, forever." One exception noted in its README: Postgres as a graph store is a demo feature, with the production version sold as a licensed product.

Does Cognee have an MCP server?

Yes, cognee-mcp, with remember, recall, forget and cognify_status tools. It runs over stdio by default, or over SSE or Streamable HTTP.

What happened to cognify?

Cognee's documentation now lists Add, Cognify and Search under "Legacy Operations." The current operations are remember, recall, improve and forget.

How much does Cognee cost?

Running it yourself is free. On October 31, 2026, Cognee Cloud listed Free ($0, one workspace and 1M tokens) and Standard ($1.00 per 1M tokens, plus $5 per additional workspace); Enterprise is a bring-your-own-cloud engagement with no listed price.

Is there a Cognee alternative for Claude Code and Codex?

Memlo is one if you want hosted memory for those agents rather than an engine to run: Claude Code, Codex and chat apps connect to the same server. See also Graphiti vs Memlo and Mem0 vs Memlo.

Sources

Checked on October 31, 2026:

• Cognee on GitHub and its license
• Cognee MCP server
• Cognee graph stores and vector stores
• Cognee documentation
• Cognee pricing
• Memlo: the setup page, how it works and terms