Letta
A platform for building stateful agents with advanced memory that learns and self-improves over time, rather than resetting between conversations.
- Actively maintained
- Well documented
- Permissive license
Overview
Letta (the team behind the original MemGPT research) provides a full platform for running stateful agents: memory that the agent itself can read, write, and reorganize over time, persistent agent state across sessions, and tooling for building, deploying, and observing these agents rather than a single library you embed elsewhere.
Letta is a fit for teams building agent products where long-running, self-editing memory is core to the product itself -- an agent that genuinely improves its understanding of a user over months of interaction -- rather than just needing a lightweight memory add-on like Mem0 (also in this catalog).
- Categories
- AI Agents & Agent Tools
- Languages
- Python
- License
- Apache-2.0
Related Projects
LangGraph
MITBuild resilient, stateful agents as graphs -- explicit nodes and edges for control flow, rather than an implicit agent loop, for reliable long-running or multi-step workflows.
Actively maintained
- AI Agents & Agent Tools
Mem0
Apache-2.0The memory layer for AI agents -- drop-in persistent memory infrastructure that remembers context across conversations and sessions.
Actively maintained
- AI Agents & Agent Tools
Composio
MITPowers 1000+ toolkits, tool search, context management and authentication so AI agents can actually take action, not just plan it.
Actively maintained
- AI Agents & Agent Tools
MCP Servers
MITOfficial reference implementations of the Model Context Protocol -- the open standard now adopted across OpenAI, Google, Microsoft and Anthropic for connecting agents to tools and data.
Actively maintained
- AI Agents & Agent Tools
Spotted an error? Suggest an edit on GitHub.