Building machines that learn.
tldr; We evaluate how well agents use the memory they have, and generate new memory from experience. Anthropic models lead overall, with the widest margin in memory generation. OpenAI models offer
Introducing trajectory, an open-source package that normalizes coding-agent sessions from @AnthropicAI Claude Code, @OpenAI Codex, @pidotdev, @LangChain deepagents, @openclaw, Letta Code, and other
tl;dr - Agents that learn like people (through experience, teaching, & self-study) will be powered by memory models: models specialized in creating and curating token-space memory that transfers
tl;dr We evaluate how well models perform for driving agents that have identity, long-lived experience, and the capability to self-evolve - both with and without additional prompting. We find that
The Context Constitution is a set of principles governing how AI agents manage context to learn from experience. We use the Context Constitution internally as the foundation of our prompting and for
Every closed foundation lab today is moving towards developing not just models, but also agents: the harness around models that enables computer use, memory, skills, subagents, and failure recovery.
We're introducing remote environments, a way to separate where you interact with a Letta agent from where it executes. Using remote environments, you can interact with Letta agents that run locally on
