Data as of Jul 25, 2026 · Based on 2,718,867 AI responses across 10,525 prompts · See how Parse measures this
Lettaletta.comLetta is a San Francisco-based AI research lab building self-improving machines and experiential agents that remember, learn, and evolve over time. Its research focuses on memory models, context management, and sleep-time compute, and it ships software such as the Letta Agent whose memory, identity, and capabilities grow with experience. The company highlights real-world case studies, is built by researchers from UC Berkeley's Sky Computing Lab, and is backed by notable investors.
Sources
letta.com shapes more of what AI says about Letta than any other source, at 18% of its citations.
agentmarketcap.ai · arxiv.org · atlan.com · python.plainenglish.io
The market map
LLM Agent Frameworks and Tooling →Where AI ranks Letta
Excerpts where Letta appeared in the AI's answer
Letta (formerly MemGPT) Focuses on OS-like memory management for agents
Letta (formerly MemGPT) — specializes in agent-managed memory rather than simply vector retrieval.
Excerpts where Letta appeared in the AI's answer
Letta (formerly MemGPT) — An SDK tailored for long-running stateful agents that require autonomous, unconstrained memory management mimicking a virtual operating system's memory hierarchy.
Letta (formerly MemGPT): Supplies an SDK geared towards long-running agents that require an OS-like virtual context management system
Excerpts where Letta appeared in the AI's answer
Letta (formerly MemGPT) — Best for OS-Style, Long-Running Autonomous Agents
Letta (formerly MemGPT) — Best for Long-Running, Stateful, Self-Editing Agents
Excerpts where Letta appeared in the AI's answer
Letta (formerly MemGPT) — Best for Deep Agent Autonomy & Self-Editing Memory
Letta (formerly MemGPT): Best for autonomous, complex, and long-running agents
