Letta Agents API is a fully-managed agents API service that transforms stateless LLMs into intelligent, stateful agents with persistent memory. Build production-ready AI applications with self-editing memory, multi-agent systems, and seamless model switching across OpenAI, Anthropic Claude (including through AWS Bedrock), Google Gemini, and Together AI - all through a simple API.
Letta is the AI operating system that transforms stateless LLMs into powerful stateful agents capable of learning, remembering, and evolving over time. Built on our team's groundbreaking MemGPT and sleep-time compute research, Letta enables developers to create AI agents with self-editing memory that maintain context across conversations, build relationships with users, and continuously improve through experience. Unlike traditional chatbots that forget everything between sessions, Letta agents possess true persistent memory - they remember past interactions, learn from conversations, and develop deeper understanding of users, your organization, and their designated tasks over time.
Highlights
Stateful agents with infinite memory - Build AI agents that remember every interaction, learn over time, and deliver personalized experiences without context window limitations through Letta's advanced memory management system.
Zero vendor lock-in with model-agnostic architecture - Switch seamlessly between OpenAI, Anthropic Claude, Google Gemini, or any supported model while preserving your agents' complete state, memories, and conversation history through our portable agent file format.
Scale to millions of agents instantly - Deploy production-ready AI applications that automatically scale from one to millions of concurrent agents with our fully-managed API service, eliminating all infrastructure complexity and DevOps overhead.
AWS Marketplace now accepts line of credit payments through the PNC Vendor Finance program. This program is available to select AWS customers in the US, excluding NV, NC, ND, TN, & VT.
This listing carries one pricing dimension, billed by the hour on a c7a.medium instance. The software itself is free, so you pay only for the compute hours you run. Your cost scales with how long you keep the instance running. There are no tiers or add-on options to choose from — usage on this single instance size is the only variable that affects what you pay.
Top-of-mind questions for buyers
What resources do I get with the c7a.medium hourly rate?
You run the software on a single c7a.medium compute instance, billed for each hour it runs. This instance size is fixed for this dimension. The hourly rate covers the software licence; standard AWS compute and storage fees for the instance apply separately.
Am I charged when the instance is stopped or paused?
Software charges accrue only while the c7a.medium instance runs. Stopping the instance halts hourly software billing. A stopped instance may still incur AWS storage fees for its attached volumes, but the software meters running time only.
What does this software do while running on the instance?
You run a self-improving AI agent whose memory and capabilities change with use. The agent manages its own context over time. The listed hourly rate applies regardless of how you use these functions on the single instance size.
Tell us how we can improve this page, or report an issue with this product.
Give us feedbackReport a problem with this product or seller
Legal
Vendor terms and conditions
Upon subscribing to this product, you must acknowledge and agree to the terms and conditions outlined in the vendor's End User License Agreement (EULA).
Content disclaimer
Vendors are responsible for their product descriptions and other product content. AWS does not warrant that vendors' product descriptions or other product content are accurate, complete, reliable, current, or error-free.
An AMI is a virtual image that provides the information required to launch an instance. Amazon EC2 (Elastic Compute Cloud) instances are virtual servers on which you can run your applications and workloads, offering varying combinations of CPU, memory, storage, and networking resources. You can launch as many instances from as many different AMIs as you need.
Version release notes
Version 0.8.8
Additional details
Usage instructions
SSH to the instance with username ubuntu over port 22 and use the following management commands.
letta-start - Start the platform
letta-stop - Stop the platform
letta-status - Show platform status
letta-logs [service] - View logs
letta-platform [command] - Full management script
"letta-platform" is a management interface for interacting with our container as a systemd service. For the full list of available configurations you can make to the underlying container, see https://docs.letta.com/guides/selfhosting
AWS Support is a one-on-one, fast-response support channel that is staffed 24x7x365 with experienced and technical support engineers. The service helps customers of all sizes and technical abilities to successfully utilize the products and features provided by Amazon Web Services.
Be the first to review this product. We've partnered with PeerSpot to gather customer feedback. You can share your experience by writing or recording a review, or scheduling a call with a PeerSpot analyst.