Skip to content

Hjarni vs Letta

Letta is a runtime for agents that manage their own evolving memory. Hjarni is a knowledge base where the human owns the content. A runtime vs a notes app.

Free to start. No credit card required.

Quick answer

Letta (formerly MemGPT) is a platform and framework for building stateful agents with OS-style tiered memory (core, recall, archival) that the agent rewrites over time. Hjarni is a Markdown knowledge base a person writes in and owns, with a built-in MCP server that ChatGPT, Claude, and other clients read and write. Pick Letta if you are a developer building and running autonomous agents that manage their own evolving memory. Pick Hjarni if you want a deliberate knowledge base where the human owns the content and AI reads it.

Hjarni Letta
Primary shape

Letta is a platform for building and running agents. Hjarni is a notes app you write in.

Knowledge base a human owns Stateful agent runtime and framework
Who owns the memory

Letta agents rewrite their own tiered memory (core, recall, archival) over time. In Hjarni, the human owns and edits the content.

You do The agent self-edits it
MCP role

Letta agents primarily use external MCP servers as tools. Hjarni is itself the MCP server your AI connects to.

Built-in server clients read and write Agents consume external MCP servers
Bring your own AI

Both are model-agnostic.

Open source and self-host

Letta is fully open-source and self-hostable via Docker, with a managed Letta Cloud too. Hjarni is hosted SaaS only, with Markdown export anytime.

Apache-2.0, self-hostable
Human notes UI

Hjarni is a notes editor a person writes in. Letta is a developer framework, not a notes app.

Full Markdown editor
Folder-level AI instructions
Built-in
Hosting region
EU (Germany) Self-host or Letta Cloud
Free tier

Hjarni's free tier is 25 notes. Letta is free to self-host; Letta Cloud pricing is not published here.

Open-source self-host
Best fit
Human-authored knowledge Agents that own their memory

An agent runtime vs a knowledge base

Letta, which grew out of the MemGPT paper, is a framework for building stateful agents. Its idea is OS-style tiered memory: a small core context, a recall layer, and an archival store, which the agent edits itself as it runs. There is also Letta Code, a coding agent. The point of Letta is the runtime, the place agents live and think and rewrite what they remember.

Hjarni is not a runtime. It is a knowledge base you write in. Notes are plain Markdown, in folders, tagged and linked, and the built-in MCP server lets ChatGPT, Claude, and other clients read and write them. The human is the author and owner; the AI is a reader and, when you allow it, a contributor.

Who owns the memory

This is the real split. In Letta, the agent owns its memory and self-edits it over time, which is the whole design. That is powerful for autonomous agents, and it means the memory is shaped by the runtime, not by a person sitting down to write.

In Hjarni, you own the content. A note changes when you change it. That predictability is the point: the next time your AI asks, it sees what you wrote, organized the way you filed it.

Letta is where an agent rewrites its own memory. Hjarni is where a person writes the knowledge down.

A note on MCP

The MCP relationship runs the opposite way for each product. Letta agents primarily consume external MCP servers as tools, calling out to capabilities you connect. There are MCP servers that expose Letta memory back to clients, but the prominent ones are community projects, so do not assume an official "expose Letta memory" server exists. Hjarni is itself the MCP server: your assistant connects to Hjarni and reads and writes your notes directly.

When Letta is the better fit

If you are a developer who wants to build and run autonomous agents that manage their own evolving memory, Letta is built for that. It is fully open-source, Apache-2.0, self-hostable via Docker, with a managed Letta Cloud if you prefer not to run it. That is the right tool when you need a runtime, not a knowledge base.

When people choose Hjarni instead

The case for Hjarni is not that Letta lacks memory. It clearly has a sophisticated memory model. The case is that some people want a deliberate knowledge base, written and owned by a human, that every AI they connect can read, rather than a runtime where the AI owns and rewrites the memory. If you want to write notes, not build agents, Hjarni is the simpler home.

When to use Letta

  • You are building and running autonomous agents
  • You want agents that manage their own evolving memory
  • You want open-source you can self-host

When to use Hjarni

  • You want a knowledge base a human writes and owns
  • You want notes that stay put until you change them
  • You want folder-level AI instructions across ChatGPT, Claude, and Cursor

Letta runs the agent. Hjarni keeps the knowledge.

Common questions

Common questions

What is Letta?

A platform and framework for building stateful agents, born from the MemGPT paper. Its agents use OS-style tiered memory (core, recall, archival) that the agent self-edits over time. There is also Letta Code, a coding agent. It is a runtime, not a notes app.

How is Letta different from Hjarni?

Who owns the memory. In Letta the agent owns and rewrites its memory as it runs. In Hjarni the human owns the content: a note changes only when you change it, and any connected AI reads it through the built-in MCP server.

Does Letta have an official MCP server for its memory?

Letta agents primarily consume external MCP servers as tools. There are community MCP servers that expose Letta memory to clients, but do not assume an official one exists. Hjarni is itself the MCP server your AI connects to.

When should I pick Letta over Hjarni?

Pick Letta when you are a developer building and running autonomous agents that manage their own evolving memory. It is fully open-source (Apache-2.0), self-hostable via Docker, with a managed Letta Cloud too. Pick Hjarni when you want a knowledge base a human writes and owns.

Can I self-host either one?

Letta is open-source and self-hostable. Hjarni is hosted SaaS only (EU, Germany), but it exports your whole knowledge base as plain Markdown anytime, so your data stays portable.

Give your AI a memory

Write once. You both remember.

Free to start. No credit card required.

Works with Claude and ChatGPT today.