Letta (MemGPT)

 »  AI Agents  »  Framework  »  Letta (MemGPT)  URL Share on

 Repository   21,736  2,295
Stateful agents with self-managed, layered memory that learn and improve over time
memory stateful self-managed-memory long-running llm-as-os mcp a2a research-backed
TypeFramework
HostingHybrid
MaturityStable
Status Maintained

Key Features

Multi-Agent Single Agent Human-in-the-Loop Streaming Async Support Short-Term Memory Long-Term Memory Shared Memory Plugin System Custom Tools MCP Protocol A2A Protocol Structured Output API Server Self-Hosted Cloud Hosted

Community Feedback

  Strengths

  • Best-in-class memory management
  • Pioneering LLM-as-OS paradigm
  • Research-backed (UC Berkeley)
  • Strong documentation
  • #1 model-agnostic on Terminal-Bench

  Weaknesses

  • Memory management complexity
  • Architecture transition (MemGPT to Letta V1)
  • Learning curve

Letta (MemGPT) Details

OrganizationLetta Inc (UC Berkeley spin-off)
Organization TypeCompany
FundingSeed
CategoryFramework
SubcategorySingle agent
DeploymentHybrid
Primary LanguagePython
RuntimePython 3.10+
LicenseApache-2.0
Commercial UsePermissive
Install Commandpip install letta
GitHub Stars21,736
GitHub Forks2,295
Release CadenceBiweekly
MaturityStable
Pricing ModelOpen core
Free TierApache-2.0 licensed core is free
Self-Hosted FreeYes
Cost Modelfree + LLM costs
Community SizeLarge (22k stars)
Community ActivityVery active
Sentiment Positive
GPU RequiredNo
Research Date2026-03-24
LLM ProvidersOpenAI, Anthropic, Google, local models
API Keys RequiredLLM provider API key

Use Cases

When to Use

Best for: Long-running agents needing persistent, self-managed memory

Avoid when: Simple stateless agent workflows or rapid prototyping

  Back to Agent Directory
Our Social Media →  
Original data from HuggingFace, OpenCompass and various public git repos.
Check out Ag3ntum — our secure, self-hosted AI agent for server management.
Release v20260328a