Letta

Build stateful AI agents with long-term memory

⭐13,000
AI Agent FrameworksFree (open-source) + Cloud
🌐 Visit Website📦 GitHub

About Letta

Letta (formerly MemGPT) provides long-term memory for AI agents. It enables stateful agents that can remember across conversations, self-edit their memory, and maintain context over extended interactions.

Features

✦Long-term memory
✦Stateful agents
✦Tool use
✦Multi-model
✦Self-editing memory

Pros & Cons

Pros

  • +Solves long-term memory problem
  • +Self-editing memory
  • +Stateful agents
  • +Multi-model support
  • +Active research backing

Cons

  • −Complex memory management
  • −Performance overhead
  • −Rebranding confusion
  • −Still experimental

Platforms

LinuxmacOSDocker

Tags

#memory#stateful#agents#open-source

Related AI Concepts

Open Source

Similar Tools

Abacus AI DeepAgent

freemium

Microsoft AutoGen

Framework for building multi-agent conversational AI

Free (open-source)

Superagentic

Full-stack agentic AI company — tools and consulting for production agents

Open-source tools + enterprise consulting

Smolagents

Hugging Face's lightweight library for building agents

Free (open-source)

📰 Featured In

All guides →
📖

How to Build Agent Memory That Actually Works

AI Tools

📖

Context Rot: Why Your AI Agent Gets Dumber Over Time (And How to Fix It)

AI Tools

📖

Context Engineering for AI Agents: The Complete Guide (2026)

AI Tools

Category

Browse all AI Agent Frameworks tools →
🤖

Need help choosing?

Compare Letta with alternatives side by side

Compare Tools →