Letta

Letta (f.k.a. MemGPT) is a framework for building stateful AI agents with long-term memory.

  • AI
Letta (f.k.a. MemGPT) is a framework for building stateful AI agents with long-term memory. It lets agents retain past conversations and personal preferences across sessions, so they can learn and improve over time instead of forgetting everything when a chat session ends.

Platform availability

Available on 0 of 10 platforms

  • Unraid (not listed)
  • TrueNAS (not listed)
  • Umbrel (not listed)
  • ZimaOS (not listed)
  • Proxmox (not listed)
  • Helm (not listed)
  • Mac (not listed)
  • Windows (not listed)
  • Linux (not listed)
  • Any server (not listed)

Health score?A 0-100 score computed nightly from four factors: maintained (40%: recent commits and steady activity), popular (25%: GitHub stars, log scale), easy to install (20%: platforms, install notes, setup guides, Docker image) and light to run (15%: minimum RAM, ARM support). When the source code has been checked, a fifth factor, security check (20%: automated scan for committed secrets, unsafe desktop-app settings, known-vulnerable dependencies and project hygiene; any serious finding caps it at 20), is added. Factors without data are left out and the rest are rescaled.

73/100 Good

Maintained
88last commit 33 days ago
Popular
8825,358 GitHub stars
Easy to install
250 install platforms, install notes, Docker image
Light to run
No data

Based on 3 of 4 factors.

GitHub stars
25,358
Open issues
0
Last commit
2026-10-18
Latest release
0.16.8
License
Apache-2.0
Activity
declining

Checked 16 days ago - source: GitHub (letta-ai/letta)

Resources & Compatibility

ARM (e.g. Raspberry Pi) is supported.

Checked 54 days ago - source: Docker Hub image tags and published docs

First-install notes

  • Install the Letta CLI first Make sure Node.js and its npm package manager are available, then run npm install -g @letta-ai/letta-code to install the CLI globally. Without the CLI, you cannot start the Letta server or manage agents from the command line. source
  • Start the server on the right port For local use, start it with letta server --backend local --listen ws://127.0.0.1:4500 so only your machine can reach it. If you change that to ws://0.0.0.0:4500 for remote access, open port 4500 and add authentication so other people on the network cannot access it. source
  • Set a real API key in the client When using the Python client, replace your-api-key in Letta(api_key=...) with a valid Letta API key before creating agents or sending messages. Without a valid key, the client cannot authenticate and its requests will fail. source
  • Install from npm Install with 'npm install -g @letta-ai/letta-code', then run 'letta server' to start the App Server for local or self-hosted agents. source

Fetched 42 days ago - each note links to its source

Alternatives (1)

  • Supermemory - Memory API and context engine for AI applications that stores knowledge from conversations, documents and connected apps, and retrieves it…

Checked 54 days ago - source: AlternativeTo

Common questions

What must I install before using the Letta CLI to start a server or manage agents?

Make sure Node.js and npm are available, then run npm install -g @letta-ai/letta-code. Without the CLI, you cannot start the Letta server or manage agents from the command line.

How can I reach a self-hosted Letta App Server from other machines?

Start the server listening on 0.0.0.0:4500 and open port 4500. Add authentication so other users on the network cannot access it.

Where is agent state stored when Letta runs as a local runtime?

In the local runtime, all agent state, including messages, memory, and provider connections, stays on-device. The desktop app uses a background App Server on your machine.

Answers sourced from docs.letta.com, docs.letta.com, docs.letta.com

Community

Collected 48 days ago - public community threads