Documentation

Community

Guides

News

Letta

API reference

Python SDK

Remove Tool From Agent

AgentType

A Block represents a reserved section of the LLM's context window which is editable. `Block` objects contained in the `Memory` object, which is able to edit the Block values.

Parameters:
    name (str): The name of the block.
    value (str): The value of the block. This is the string that is represented in the context window.
    limit (int): The character limit of the block.
    template (bool): Whether the block is a template (e.g. saved human/persona options). Non-template blocks are not stored in the database and are ephemeral, while templated blocks are stored in the database.
    label (str): The label of the block (e.g. 'human', 'persona'). This defines a category for the block.
    description (str): Description of the block.
    metadata_ (Dict): Metadata of the block.
    user_id (str): The unique identifier of the user associated with the block.

Description

Label of the block (e.g. 'human', 'persona').

Label

Limit

Metadata 

Name

Whether the block is a template (e.g. saved human/persona options).

Template

The unique identifier of the user associated with the block.

User Id

Value

Block

Embedding model configuration. This object specifies all the information necessary to access an embedding model to usage with Letta, except for secret keys.

Attributes:
    embedding_endpoint_type (str): The endpoint type for the model.
    embedding_endpoint (str): The endpoint for the model.
    embedding_model (str): The model for the embedding.
    embedding_dim (int): The dimension of the embedding.
    embedding_chunk_size (int): The chunk size of the embedding.
    azure_endpoint (:obj:`str`, optional): The Azure endpoint for the model (Azure only).
    azure_version (str): The Azure version for the model (Azure only).
    azure_deployment (str): The Azure deployment for the model (Azure only).

Azure Deployment

Azure Endpoint

Azure Version

Embedding Chunk Size

Embedding Dim

The endpoint for the model (`None` if local).

Embedding Endpoint

Embedding Endpoint Type

Embedding Model

EmbeddingConfig

Configuration for a Language Model (LLM) model. This object specifies all the information necessary to access an LLM model to usage with Letta, except for secret keys.

Attributes:
    model (str): The name of the LLM model.
    model_endpoint_type (str): The endpoint type for the model.
    model_endpoint (str): The endpoint for the model.
    model_wrapper (str): The wrapper for the model. This is used to wrap additional text around the input/output of the model. This is useful for text-to-text completions, such as the Completions API in OpenAI.
    context_window (int): The context window size for the model.

Context Window

Model

Model Endpoint

Model Endpoint Type

Model Wrapper

LLMConfig

Represents the in-context memory of the agent. This includes both the `Block` objects (labelled by sections), as well as tools to edit the blocks.

Attributes:
    memory (Dict[str, Block]): Mapping from memory block section to memory block.

Mapping from memory block section to memory block.

Memory

Jinja2 template for compiling memory blocks into a prompt string

Prompt Template

Location

Message

Error Type

ValidationError

bearerAuth

Representation of an agent's state. This is the state of the agent at a given time, and is persisted in the DB backend. The state has all the information needed to recreate a persisted agent.

Parameters:
    id (str): The unique identifier of the agent.
    name (str): The name of the agent (must be unique to the user).
    created_at (datetime): The datetime the agent was created.
    message_ids (List[str]): The ids of the messages in the agent's in-context memory.
    memory (Memory): The in-context memory of the agent.
    tools (List[str]): The tools used by the agent. This includes any memory editing functions specified in `memory`.
    system (str): The system prompt used by the agent.
    llm_config (LLMConfig): The LLM configuration used by the agent.
    embedding_config (EmbeddingConfig): The embedding configuration used by the agent.

AgentState

HTTPValidationError

Letta is an AI platform for building stateful LLM applications.

Overview

Welcome to Letta

Quickstart

Developer quickstart

Install and setup Letta to run the server & ADE

Run with pip

Run Letta server with pip

Run with Docker

Run Letta server with Docker

OpenAI

OpenAI proxy

OpenAI-compatible endpoint

Anthropic

Groq

Together

Google AI (Gemini)

Ollama

vLLM

Azure OpenAI

Letta concepts

Key concepts

MemGPT concepts

MemGPT

Tools

Messaging agents

Message streaming

Streaming support

Understanding the concept of LLM memory management introduced in MemGPT

Memory management

Understanding memory management

Learn how to create custom memory classes

Custom memory

Creating custom memory classes

Get Tool

Delete Tool

Update Tool

Get Tool Id

Get a list of all tools available to agents created by a user

List All Tools

Create Tool

Get Source

Delete Source

Update the name or documentation of an existing data source.

Update Source

Get Source Id By Name

List Sources

Create Source

Attach a data source to an existing agent.

Attach Source To Agent

Detach a data source from an existing agent.

Detach Source From Agent

Upload File To Source

List all passages associated with a data source.

List Passages

List paginated files associated with a data source.

List Files From Source

Delete File From Source

List all agents associated with a given user.
This endpoint retrieves a list of all agents and their configurations associated with the specified user ID.

List Agents

Create a new agent with the specified configuration.

Create Agent

Retrieve the context window of a specific agent.

Get Agent Context Window

Get Agent State

Delete Agent

Update Agent

Add Tool To Agent

Get the sources associated with an agent.

Get Agent Sources

Retrieve the messages in the context of a specific agent.

Get Agent In Context Messages

Retrieve the memory state of a specific agent.
This endpoint fetches the current memory state of the agent identified by the user ID and agent ID.

Get Agent Memory

Update the core memory of a specific agent.
This endpoint accepts new memory contents (human and persona) and updates the core memory of the agent identified by the user ID and agent ID.

Update Agent Memory

Retrieve the summary of the recall memory of a specific agent.

Get Agent Recall Memory Summary

Retrieve the summary of the archival memory of a specific agent.

Get Agent Archival Memory Summary

Retrieve the memories in an agent's archival memory store (paginated query).

Get Agent Archival Memory

Insert a memory into an agent's archival memory store.

Insert Agent Archival Memory

Delete a memory from an agent's archival memory store.

Delete Agent Archival Memory

Get Agent Messages

Process a user message and return the agent's response.
This endpoint accepts a message from a user and processes it through the agent.
It can optionally stream the response if 'stream_steps' or 'stream_tokens' is set to True.

tools

sources

agents

models

blocks

jobs

health

Remove Tool From Agent

Authorizations

Headers

Path Parameters

Response