Letta agent file. You can interact with the .
Letta agent file. a “concious” and a “subconcious”), or agents with more advanced memory types (e. GitHub - letta-ai/agent-file: Agent File (. Our chatbot webapp template showcases powerful core features of Letta. Letta Formerly known as MemGPT, Letta is an open-source framework designed for building stateful LLM applications. Python 898 81 letta-chatbot-example Public Feb 13, 2025 路 Although RAG provides a way to connect LLMs and agents to more data than what can fit into context, traditional RAG is insufficient for building agent memory. Letta acts as the host, creating clients that connect to external servers. Jul 24, 2025 路 The platform for stateful agents. The Letta Filesystem represents documents as folders and files (containing parsed contents) to the agent, and provides the agent with filesystem-like tools (e. 馃懢 Letta is an open source framework for building stateful agents with advanced reasoning capabilities and transparent long-term memory. open file, grep) to interact with file contents. Letta and MemGPT introduced the concept of agentic context engineering, where the context window engineering is done by one or more AI agents. While Letta can be self-hosted, Letta Cloud eliminates all infrastructure management, server optimization, and system administration so you can focus entirely on building agents. Build AI agents with long-term memory, advanced reasoning, and custom tools inside a visual environment using the Agent Development Environment, or with Python and Node. They’re ideal for real-time applications like voice interfaces where latency matters more than context retention. In Letta, you can create special sleep-time agents that share the memory of your primary agents, but run in the background and can modify the memory asynchronously. This tool enables developers to package agent components—such as memory, prompts, configurations, and settings—into a single, portable file. Building the #1 open source terminal-use agent with Letta shows that general memory management in Letta provides effective building blocks for better and more performant agents beyond long-running chatbots. To enable access to these LLM API providers, set the appropriate environment variables when you use docker run: Are you developing an application on Letta using ChatGPT, Cursor, Loveable, or another AI tool? Use our pre-made prompts to teach your AI how to use Letta properly. Share, checkpoint, and version control agents across compatible frameworks. While basic memory might simply involve recalling previous interactions, advanced memory systems enable agents to learn and improve over time, adapting their behavior based on accumulated experience. Reason Cookbooks and Tutorials Learn how to build with Letta using tutorials and pre-made apps Sep 23, 2024 路 A startup called Letta has just emerged from stealth with tech that helps AI models remember users and conversations. Use the workflow_agent agent type for structured, sequential processes where you need deterministic execution paths. You can create a memory block that persists information in-context You can create a file which the agent can read segments of and search You can write to archival memory for the agent to later query via built-in tools You can use an external DB (e. Static defined agent state Define your agent state in the default-agents. Creates a template from an agent. json configuration files. You can interact with the Oct 14, 2024 路 We are deprecating the letta configure and letta quickstart commands, and the the use of ~/. Introducing “stateful agents”: AI systems that maintain persistent memory and actually learn during deployment, not just during training. The platform for stateful agents. It seems to me that it's not about ollama, since everything works with llama_index and langchain Install pip install letta Agent setting from l This notebook is a tutorial on how to use Letta's LocalClient. In Letta, the agent type memgpt_agent implements the original agent architecture from the MemGPT research paper, which includes a set of base tools: send_message: required for sending messages to the user core_memory_append and core_memory Workflows execute predefined sequences of tool calls with LLM-driven decision making. Build with Letta Learn how to build and deploy stateful agents Jul 31, 2025 路 The Letta ADE is a graphical user interface for creating, deploying, interacting and observing with your Letta agents. 3 days ago 路 Today we're announcing Letta Filesystem, which provides an interface for agents to organize and reference content from documents like PDFs, transcripts, documentation, and more. Sep 23, 2024 路 Letta will continue to develop and maintain the MemGPT open source software (permissively licensed under Apache 2. Download the Letta open source framework on GitHub Letta enables you to build and deploy stateful AI agents that maintain memory and context across long-running conversations. Opens a specific file for a given agent. The file will be removed from the agent's working memory view. - letta-ai/letta. Learn how to build and deploy stateful agents Get started → {/* Main Content */} Create your first stateful agent in a few minutes Learn how to use the Agent Development Environment (ADE) Integrate Letta into your application with a few lines of code Connect Letta agents to tool libraries via Model Context Protocol (MCP) Learn how to build with Letta using tutorials and pre-made apps Take You can create custom tools in Letta using the Python SDK, as well as via the ADE tool builder. Agents continue to exist and maintain state even when your application isn’t running, with computation happening on the server and all memory, context, and tool connections handled by the Letta server. This endpoint marks a specific file as open in the agent's file state. For your agent to call a tool, Letta constructs an OpenAI tool schema (contained in json_schema field) from the function you define. g. This agent is characterized by its advanced memory management capabilities, enabling it to maintain context and personalized information across conversations Letta agents live inside the Letta server, which persists them to a database. af) Scheduling Voice Agents Tool Use MemGPT agents are equipped with memory-editing tools that allow them to edit their in-context memory, and pull external data into the context window. This file contains the initial state of your agents, including the LLM model, user profile, agent persona, and other 3 days ago 路 With Letta, agent developers can rapidly specialize agents for specific tasks, by focusing on building the right prompts, tools, and environment. Workflows are stateless by default but can branch and make decisions based on tool outputs and LLM reasoning. At its core, the ADE makes the once-opaque world of context windows and agent reasoning visible and manageable for developers. Purpose and Scope The Letta Agent File repository demonstrates how to implement different types of AI agents using tool-based architectures, rules systems, and memory management patterns. You can use Letta to create powerful LLM agents that can reason about their past interactions, learn from them, and improve their behavior over time. 馃懢 Letta is an open source framework for building stateful LLM applications. This endpoint marks a specific file as closed in the agent's file state. Editing Memory Blocks via the Letta API Unlike ephemeral memory in many LLM frameworks, Letta's memory blocks are individually persisted in the DB, with a unique `block_id` to access them via the API and Agent Development Environment (ADE). Letta v0. May 4, 2025 路 Letta is a model-agnostic, open-source framework that allows developers to build, deploy, and manage stateful AI agents. 3 days ago 路 Today we're announcing Letta Filesystem, which provides an interface for agents to organize and reference content from documents like PDFs, transcripts, documentation, and more. af) is an open standard file format for serializing stateful agents. Apr 17, 2025 路 Letta: The OS for Agents Letta is the operating system for AI agents, managing state, context, and execution so developers can build intelligent applications that actually learn and improve with experience. Unlike the RESTClient which connects to a running agents service, the LocalClient will run agents on your local machine, so does not require connecting to a service. vector DB, RAG DB) to store data, and make the data accessible to your agent via tool calling (e. When attached, your agent automatically gains file tools to search and access the content. Ensure your agent is configured with a multi-modal capable model. While most frameworks are librariesthat wrap model APIs, Letta provides a dedicated servicewhere agents live and operate autonomously. Letta is fundamentally different from other agent frameworks. - letta-ai/age The Letta server can be connected to various LLM API backends (OpenAI, Anthropic, vLLM, Ollama, etc. The Letta framework also allow you to make agent architectures beyond MemGPT that differ significantly from the architecture proposed in the research paper - for example, agents with multiple logical threads (e. Jul 7, 2025 路 What is Agent Memory? Agent memory is what and how your agent remembers information over time. The search technique is pulled from this academic paper on DeepRAG, although query decomposition is a well known technique in general. This allows any Letta agent to gain blockchain-verified compliance registration with minimal friction. This endpoint is only available on Letta Cloud. af): An open file format for serializing stateful agents with persistent memory and behavior. Letta agents live inside the Letta server, which persists them to a database. task memory). sending a message to another agent). These tools include memory management tools (for reading and writing to memory blocks), file editing tools, multi-agent tools, and general utility tools like web search and code execution. Letta (formerly MemGPT) is the stateful agents framework with memory, reasoning, and context management. Letta can either parse this automatically from a properly formatting docstring, or you can pass in the schema explicitly by providing a Pydantic object that defines Closes a specific file for a given agent. 1 adds improved support for popular external tool providers - you can now use external tool libraries (Composio, LangChain, and CrewAI) with Letta agents! To active support for external tool providers in Letta, install Letta with the external-tools library: pip install 'letta[external-tools]' If set to True, the agent will not remember previous messages (though the agent will still retain state via core memory blocks and archival/recall memory). You can think of sleep-time agents as a special form of multi-agent architecture, where all agents in the system share one or more memory blocks. This open Managing data sources in the ADEThe Data Sources panel in the ADE allows you to connect external files to your agent. af) 馃懢 – solving one of the biggest challenges in AI development by making stateful agents truly portable and shareable for the first time. Today we're launching Agent File (. the Letta agent framework (you can see more with our OSS announcement). json file. I would like to use vllm server with streaming support. AstraSync AF Bridge Bridge for importing Letta Agent Files (. js SDKs. Export the serialized JSON representation of an agent, formatted with indentation. Jul 24, 2025 路 In this video, Cameron walks you through how to use Letta Filesystem to create folders of PDFs, text files, and other documents Letta agents can use. Letta Cloud is our fully-managed service for stateful agents. 0), though the package names will be shifted to Letta to make clear the distinction between MemGPT agents vs. Letta provides a set of pre-built tools that are available to all agents. - letta-ai/letta Letta is built by the creators of MemGPT, a research paper that introduced the concept of an “LLM Operating System” for memory management. The Letta Agent File repository provides a collection of agent implementations that demonstrate different approaches to building AI agents using tool-based architectures, memory systems, and rule-based workflows. These agents can: Maintain long-term memory across interactions. Returns a list of file names that were closed due to LRU eviction. For example, if you're running a Letta server to power an end-user application (such as a customer support chatbot), you can use the ADE to test, debug, and observe the agents in your server. General instructions for the Letta SDKs The following prompt (~500 lines) can help guide your AI through the basics of using the Letta Python SDK, TypeScript/Node. Unlike traditional frameworks that rely on stateless APIs, Letta functions as a complete agent operating system where agents run as persistent services that maintain state across Nov 13, 2024 路 Is your feature request related to a problem? Please describe. they say that tools output is openai compatible, see: vercel/ai#2231 https://d Low-latency agents optimize for minimal response time by using a constrained context window and aggressive memory management. The base agent design in Letta is a MemGPT-style agent, which means it inherits the core principles of: Dec 18, 2024 路 Bug description Letta does not see the ollama server. This endpoint fetches the current memory state of the agent identified by the user ID and agent ID. You can also use the ADE as a general chat interface to interact with your Letta agents. Develop agents that truly learn and evolve from interactions without starting from scratch each time. To associate agents you create in Letta with your users, you can first create an Identityobject with the user’s unique ID as the identifier_keyfor your user, and then specify the Identityobject ID when creating an agent. The repository contains complete, functional agent implementations that can be studied, modified, or used as reference for building new agents. When you use the /messages/stream route, stream_steps is enabled by default, and the response to the POST request will stream back as server-sent events (read Architecture MCP uses a host-client-server model. Streaming agent steps When you send a message to the Letta server, the agent may run multiple steps while generating a response. MCP) The Agent File (. 4. Letta agents support image inputs, enabling richer conversations and more powerful agent capabilities. This endpoint retrieves a list of all agents and their configurations associated with the specified user ID. You can interact with the Letta agents inside your Letta server via the REST API + Python / Typescript SDKs, and the Agent Development Environment (a graphical interface). Letta agents live inside the Letta server, which Multi-modal features require compatible language models. Agent File (. Copy-paste the following into Letta (formerly MemGPT) is the stateful agents framework with memory, reasoning, and context management. Instead of sitting idle between tasks, AI agents can now use their "sleep" time to process information and form new connections by rewriting their memory state. Apr 6, 2025 路 Letta is an agent framework that has built-in self editing memory and built-in tooling for editing the behavior of the agent, including adding new tools. Agent File Introducing Agent File (. af) is an open standard format designed for serializing stateful AI agents, initially created for the Letta framework. List all agents associated with a given user. letta/config for specifying the default LLMConfig and EmbeddingConfig, as it prevents a single letta server from being able to run agents with different model configurations concurrently, or to change the model configuration of an agent without re Apr 21, 2025 路 Sleep-time compute is a new way to scale AI capabilities: letting models "think" during downtime. af) into the AstraSync compliance platform. The file will be included in the agent's working memory view. ). You can use Letta to build stateful agents with advanced reasoning capabilities and transparent long-term memory. Retrieve the memory state of a specific agent. The Letta framework is white box and model-agnostic. Nov 7, 2024 路 Today we're announcing Letta Filesystem, which provides an interface for agents to organize and reference content from documents like PDFs, transcripts, documentation, and more. If true, attaches the Letta multi-agent tools (e. The response format used by the agent when returning from send_message. Letta supports three MCP transport types depending on your deployment and use case. af): An open file format for serializing stateful AI agents with persistent memory and behavior. The Letta server can be connected to various LLM API backends (OpenAI, Anthropic, vLLM, Ollama We would like to show you a description here but the site won’t allow us. Each server exposes tools, resources, or prompts through the standardized MCP protocol. This provides a way for developers to directly modify parts of their agent’s context window. Jan 15, 2025 路 Watch a tutorial video on the Letta ADE on our YouTube channel! We're excited to announce the Agent Development Environment (ADE), a visual development environment that brings unprecedented transparency to agent design and debugging. MemGPT Agent Relevant source files Purpose and Scope This document provides technical documentation on the MemGPT Agent, a stateful conversational AI agent implementation based on the MemGPT framework within the Letta Agent ecosystem. Originally designed for the Letta framework, Agent File provides a portable way to share agents with persistent memory and behavior. Closes a specific file for a given agent. Letta’s MCP integration connects your agents to external tools and data sources without requiring custom integrations. This tutorial will cover the basics of creating an agent, interacting with an agent, and understanding the agent's state and memories. In Letta, agents are able to manage their own context window (and the context window of other agents!) using special memory management tools. js SDK, and Vercel AI SDK integration. If set to True, the agent will not remember previous messages (though the agent will still retain state via core memory blocks and archival/recall memory). Connect Letta agents to tools over Model Context Protocol (MCP) Letta no longer supports legacy . New strip_messages field in Import Agent API The Import Agent API now supports a new strip_messages field to remove messages from the agent’s conversation history when importing a serialized agent file. Multi-Modal Multi-Agent Multi-User (Identities) Agent File (. Not recommended unless you have an advanced use case. Letta agents automatically come with a set of memory management tools that allow agents to search previous messages by text or data, write memories, and edit the agent’s own context window (you can read more here). The Lett Retrieve the memories in an agent’s archival memory store (paginated query). Use the ADE or API/SDK. For example, an agent may run a search query, then use the results of that query to generate a response.
agujme mlv vjhu gbct lfaoe rtuvfd bwphxe xyg eibie wntjdr