Letta
Letta (f.k.a. MemGPT) is a framework for building stateful AI agents with long-term memory.
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
Variants
Alternatives (1)
- Supermemory - Memory API and context engine for AI applications that stores knowledge from conversations, documents and connected apps, and retrieves itβ¦
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
- Reddit (reddit)
- Letta Developer Community (forum)
- Latest Community topics - Letta Developer Community - forum.letta.com (forum)
Collected 48 days ago - public community threads