Letta

Letta is an open source agent framework that provides stateful AI memory to build persistent agents capable of learning from experience.

Letta is a platform for building stateful agents that maintain long-term memory, allowing them to learn and improve over time. Developers can utilize the Agent SDK to embed these agents into applications or use the Letta app and CLI for personal workflows. The architecture supports integrating custom skills and recurring schedules to automate tasks. Users can deploy agents via managed Letta Cloud or run the runtime locally. It is designed for developers creating coding assistants, automated coworkers, and personal tools.

Letta at a glance

Best for: Software engineer
Runs on: CLI, Slack, self-hosted
Pricing: Free: $0/month (limited); Pro: $20/month; Teams Pro: $20/seat/month; API: $0.10/active agent/mo plus $0.00015/sec tool execution; Enterprise: custom.
Released: October 2023
Category: AI agent tool
Website: docs.letta.com

Key features

Stateful memory management for agent persistence
Agent SDK for application integration
Customizable skills system for workflow extension
Schedules for automating recurring prompts
Multi-platform support for Slack, Telegram, and Discord
Local runtime execution via Letta App Server
Managed deployment through Letta Cloud

Use it for

Automating recurring work tasks
Building AI coding assistants
Creating persistent personal assistants
Deploying AI coworkers in messaging channels

Alternatives to Letta

memU — It gives every connected agent a shared and inspectable memory layer.
Tigerless Agent Memory — Maintains agent history in portable, greppable Markdown files without API keys.
MemOS — It delivers ultra-persistent memory, hybrid-retrieval, and token savings.
MyAgent — It features built-in permission controls, long-term memory, and skill self-evolution.
OpenSquilla — Maximizes AI intelligence density while minimizing token costs through smart routing.
OpenViking — Unifies agent memory, knowledge RAG, and skills into a single directory structure.
Supermemory — Integrates long-term memory into AI coding agents with a single prompt.
Agentmemoryrepo — Enables agents to navigate memories like a wiki and share knowledge across sessions.
Memori — Reduces LLM token usage and spend by over 97% with targeted recall.
EverOS — It enables agents to learn from experience and improves over time without manual coding.
ReMe — It enables local-first, file-native long-term memory sharing across multiple AI agents.
Beeuni — Agents retain context and improve skills over time through a controlled review process.

All Letta alternatives