Updated

Letta

Platform and API for stateful agents that keep long-term memory and improve with use, from the creators of MemGPT, plus Letta Code, a terminal agent built on it.

3 makers' products we track use Letta. 1 open-source projects also declare it in their code.

Open sourceSelf-hostableFree tierPaid from $20/mo

Profile checked

Ask your AI about this, with this page as the source:ChatGPT ↗Claude ↗Perplexity ↗

How makers use Letta

Makers' products we track use Letta for agent memory (3).

What Letta is used for

Agent Memory · Stateful agent platform →Building the agent itself on a platform where long-term memory, self-editing context and tools are built in.Trade-off: Memory comes with Letta's agent model and API, so moving to another framework later means rebuilding; the company now leads with its own coding agent.
Letta3Mem02Supermemory2Graphiti1Zep0Makers' products on each, for agent memory
AI SDK & Agent Framework · Broad framework →Building stateful agents on a server and API where long-term memory and self-editing context are built in.Trade-off: Agents live in Letta's server model, so leaving means rebuilding; the company now leads with its own coding agent.
LangChain66Mastra24LlamaIndex16AI SDK5CrewAI3Letta0Makers' products on each, for AI SDK & agent framework

When to use Letta, and when not

Pick Letta if

  • You'd rather build the agent on a platform where memory is built in. Agent Memory →

Something else fits better if

  • You want to add user memory to an existing agent with a few API calls and a free tier to start: Mem0 or Supermemory.
  • Memory must stay on your own infrastructure, built as a graph over your data: Cognee or Graphiti.
  • Your agent needs dated facts that change over time, from business data as well as chats: Zep or Graphiti.
  • You want memory as plain Markdown files you can read, edit and share across your own AI tools: Basic Memory.
  • Your app is TypeScript and you want agents without adding a Python service: AI SDK or Mastra.
From each decision's guide, written from what makers' products chose and why.

Who uses it

3 makers' products, each linked to the source that shows it, and 1 open-source projects that declare it in their code.

Customer story 3Declared in code 1How evidence is collected →

Open source: a project that declares Letta as a dependency in its public code — verifiable, but not necessarily a live product.

Further reading

Questions makers ask about Letta

Is Letta good for production?

3 makers' products we track run on it.

What is Letta used for?

Platform and API for stateful agents that keep long-term memory and improve with use, from the creators of MemGPT, plus Letta Code, a terminal agent built on it. Makers' products we track use it for agent memory (3).

When should you use Letta?

You'd rather build the agent on a platform where memory is built in (agent memory). Something else fits better if: you want to add user memory to an existing agent with a few API calls and a free tier to start, pick Mem0 or Supermemory; Memory must stay on your own infrastructure, built as a graph over your data, pick Cognee or Graphiti.

Is Letta free?

Yes — there is a free tier a small product can run on; paid use starts at $20/mo. source ↗

Is Letta open source or self-hostable?

Open source, and you can self-host it. source ↗

Can AI coding agents work with Letta?

No llms.txt, official MCP server or CLI found yet.

Who uses Letta?

3 makers' products we track, each with a source, and 1 open-source projects declare it in their code. source ↗