Open Source Alternatives LogoOpen Source Alternatives
AlternativesBlogAdvertise
Open Source Alternatives LogoOpen Source Alternatives

Stay Updated

Subscribe to our newsletter for the latest news and updates about Alternatives

Open Source Alternatives LogoOpen Source Alternatives

Handpicked Open Source Alternatives to Paid Softwares

Product
  • Search
  • Categories
  • Tag
  • Sign In
Resources
  • Blog
  • Collection
  • Submit
  • Advertise your tool
Company
  • Privacy Policy
  • Terms of Service
  • Refund Policy
  • Sitemap
Copyright © 2026 All Rights Reserved.
Home/Categories/LLMOps & AI Tooling/Letta
icon of Letta

Letta

Open source alternative to Vellum AI, Google Vertex AI Agent Builder and Kore.ai Agent Platform

Letta is an open source AI agent memory framework that gives LLM agents persistent, self-editing long-term memory so they remember users and context across sessions. Apache 2.0.

2.5K starsTypeScriptApache-2.0
image of Letta
Contents
  1. 01Who Letta is for

Repository

Stars
2.5K
Forks
252
License
Apache-2.0
Latest
v0.25.8
Last commit
21 days ago
Last verified
May 13, 2026
Repo
letta-ai/letta-code ↗
Active this month
Visit websiteGitHub repo
  • 02The problem it solves
  • 03How it solves it
  • 04Strengths and trade-offs
  • 05Tech stack
  • 06FAQ
  • 07Similar open-source tools
    • TL;DR

    Letta is an AI agent memory framework for building stateful agents that remember users, tools, and context across sessions. It replaces stateless prompt chains and basic RAG wrappers for teams that need persistent agent behavior. Apache-2.0 licensed repository.Apache-2.0 · TypeScript · 2.5K stars · Active this month

    who it's for

    Who Letta is for#

    AI teams building personal assistants

    Use Letta when an assistant needs to remember preferences, history, and working context across conversations.

    Skip if:

    A stateless chatbot with fresh retrieval is enough for your use case.

    Developers prototyping agent memory

    Use Letta to test how memory changes agent behavior before building custom memory infrastructure.

    Skip if:

    You cannot store user memory because of policy or compliance limits.

    the problem

    The problem it solves#

    how Letta solves it

    How it solves it#

    Persistent agent memory

    Letta focuses on agents that retain and update memory across sessions, helping them behave consistently over time.

    Framework for stateful agents

    Developers can build agent behavior around memory, tools, and context rather than a single stateless model call.

    LLMOps-oriented design

    The project fits AI engineering workflows where memory, observability, and control matter for production agent behavior.

    strengths · trade-offs

    Strengths and trade-offs#

    Strengths

    • Treats memory as core infrastructureLetta is useful when agent memory needs explicit management instead of hidden prompt tricks.
    • Better fit for long-running agentsPersistent memory helps assistants, copilots, and workflow agents carry useful context between sessions.

    Trade-offs

    • -Memory needs governancePersistent memory can store sensitive or stale information. Teams need clear retention, editing, and deletion policies.
    tech stack · detected from GitHub

    What it's built on#

    Languages
    TypeScript
    Frameworks
    React
    frequently asked

    FAQ#

    What is Letta used for?

    Letta is used to build AI agents with persistent memory and stateful behavior across sessions.

    Does Letta replace RAG?

    Letta does not simply replace RAG. It focuses on agent memory, which can work alongside retrieval when agents need both remembered context and external knowledge.

    Who should use Letta?
    also worth a look

    Similar open-source tools#

    OpenMolt

    OpenMolt

    Build programmatic AI agents in Node.js, open source

    34TypeScriptMIT
    Caveman

    Caveman

    Token-slashing caveman-speak for cheaper, faster AI code.

    67.5KJavaScriptMIT
    Skills Manager

    Skills Manager

    Universal skills manager for Claude Code and AI coding agents

    57TypeScriptMIT
    AI-Flow

    AI-Flow

    Visually chain AI models and APIs into automated pipelines

    283TypeScriptMIT
    Botpress

    Botpress

    Visual chatbot builder with LLM integration and live deployment

    22.6KJavaScriptMIT
    jcode

    jcode

    Next-gen coding agent harness for efficient workflows

    6KRustMIT

    Additional details

    Language
    TypeScript
    Open issues
    162
    Contributors
    26
    First release
    2025

    Categories

    LLMOps & AI ToolingAI & Machine LearningDeveloper Tools

    Tags

    AI AgentsLLMLLMOpsDeveloper FrameworkDeveloper ToolsRAG

    Most LLM applications forget context unless developers keep pushing conversation history, summaries, or retrieval results back into the prompt. That creates brittle behavior and makes agents feel inconsistent across sessions.

    Long-running agents need memory as a managed product concern. Teams need to decide what the agent can remember, update, retrieve, and expose instead of treating memory as an accidental prompt side effect.

    Letta is best for AI teams building assistants or agents where long-term context changes the product experience.