Platform for stateful AI agents with advanced memory, continual learning, and application APIs
- Stars23.3k
- Forks2.5k
- Open Issues55
Apache-2.0
- Python
- Go
- Shell

About Letta
Letta is a platform for building AI agents with advanced memory that can learn and self-improve over time. Formerly MemGPT, it is aimed at stateful agents that remember, learn continuously, and can be built into applications through an agents API.
Letta Code runs agents locally in a terminal, where an agent can help with coding and other tasks on the computer. It supports skills and subagents, with pre-built skills and subagents for advanced memory and continual learning. Letta is model-agnostic, and the API is available with Python and TypeScript SDKs.
Letta is open source and built by over a hundred contributors from around the world. The CLI requires Node.js 18+ and installs as an npm package, while the API supports TypeScript and Python application integration. Related Letta services such as the hosted endpoint are covered by Letta terms and privacy policy.
Key features
- Stateful agents with advanced memory
- Local terminal agents through Letta Code
- Skills and subagents for agent workflows
- Pre-built memory and continual learning skills
- Agents API with Python and TypeScript SDKs
Details
- First released
- 2023
- Former name
- MemGPT
- CLI
- Letta Code
- Runtime
- Node.js 18+
- SDKs
- Python · TypeScript
- Models
- Model-agnostic
