letta
Create LLM agents with long-term memory and custom tools
Decision gist · record as of 2026-08-14
Yes, if you need a production-grade agent framework with memory management and tool integration. The active maintenance, permissive license, and low install friction support adoption. However, the large dependency tree and requirement for an external API key (for the cloud API) or Node.js 18+ (for the CLI) may be friction points. Start with letta-client if you only need the Python SDK; the full letta package is for running the backend service.AI-flagged interpretation of the facts on this page — verify before relying
Before you install
- Requires a Letta API key (from app.letta.com) and Python 3.11 or later.
- The CLI tool also requires Node.js 18+ if using the Letta Code command-line interface.
- Low install friction with a pure-Python wheel distribution.
License · maintenance · safety
Apache License (permissive) — Licensed under Apache License with permissive treatment, allowing commercial and private use with minimal restrictions.
last release 2026-05-14 (92 days)
0 known vulnerabilities (OSV.dev, 2026-08-14) · 148,733 downloads/mo, #11,020 on PyPI
Alternatives
Verify before relying
pip install letta
from letta_client import Letta
import os
client = Letta(api_key=os.getenv("LETTA_API_KEY"))
agent = client.agents.create(
model="openai/gpt-5.2",
memory_blocks=[{"label": "human", "value": "Name: Alice"}],
tools=["web_search"]
)
response = client.agents.messages.create(
agent_id=agent.id,
input="What do you know about me?"
)- Whether all 69 runtime dependencies are required for basic usage or if many are optional for specific integrations.
- Performance characteristics and latency when managing agent state and memory across requests.
- Compatibility details with specific LLM providers beyond the examples shown (Anthropic, OpenAI).
What it is and what it does
Letta is a framework for creating AI agents that maintain long-term memory and can be customized with tools and skills. It offers two main entry points: a local CLI tool (Letta Code, installed via npm) that runs agents on your machine, and a Python API (letta-client) for embedding stateful agents into applications. The framework handles agent state management, memory persistence, and tool integration, allowing agents to learn and adapt over time.
The package itself (letta) is the backend service and core framework, while letta-client is the Python SDK for API interaction. It supports multiple LLM models and is model-agnostic, though the documentation recommends specific models for optimal performance. The dependency footprint is substantial—69 runtime packages including integrations with Anthropic, OpenAI embeddings, LlamaIndex, and observability tools like Datadog—reflecting its role as a full-featured agent platform rather than a lightweight library.
Use it for
- Build conversational AI assistants that remember user context and preferences across sessions.
- Create autonomous agents that can use web search, fetch webpages, and other tools to complete tasks.
- Integrate stateful LLM agents into existing applications via the Python SDK with persistent memory.
- Develop self-improving agents that learn from interactions and adapt their behavior over time.
- Run local AI agents in the terminal for coding assistance and task automation using Letta Code.
Worth the install?
AI-flagged interpretation of the facts on this page. Verify before relying on it.
Yes, if you need a production-grade agent framework with memory management and tool integration.
The active maintenance, permissive license, and low install friction support adoption. However, the large dependency tree and requirement for an external API key (for the cloud API) or Node.js 18+ (for the CLI) may be friction points. Start with letta-client if you only need the Python SDK; the full letta package is for running the backend service.
Install
letta on PyPI
Before you install
Low install friction with a pure-Python wheel distribution. Actively maintained with a recent release. Requires Python 3.11 or later and brings in a large dependency tree (69 runtime packages including anthropic, llama-index, and datadog), which may add setup complexity in constrained environments.
Requires a Letta API key (from app.letta.com) and Python 3.11 or later. The CLI tool also requires Node.js 18+ if using the Letta Code command-line interface.
License in practice
Licensed under Apache License with permissive treatment, allowing commercial and private use with minimal restrictions.
Quickstart
pip install letta
from letta_client import Letta
import os
client = Letta(api_key=os.getenv("LETTA_API_KEY"))
agent = client.agents.create(
model="openai/gpt-5.2",
memory_blocks=[{"label": "human", "value": "Name: Alice"}],
tools=["web_search"]
)
response = client.agents.messages.create(
agent_id=agent.id,
input="What do you know about me?"
)Verify before relying
- Whether all 69 runtime dependencies are required for basic usage or if many are optional for specific integrations.
- Performance characteristics and latency when managing agent state and memory across requests.
- Compatibility details with specific LLM providers beyond the examples shown (Anthropic, OpenAI).
Package facts
| License | Apache License permissive |
| Python support | Supports the current Python release <3.14,>=3.11 |
| Install friction | Low. Pure-Python wheel |
| Runtime dependencies | 69 packagesaiofilesaiomultiprocessalembicanthropicapschedulerasync-lrublackbrotlicertificlickhouse-connectcoloramadatadogdatamodel-code-generatorddtracedemjson3docstring-parserexa-pyfakerfastmcpgoogle-genaigrpcio-toolsgrpciohtml2texthttpx-ssehttpxletta-clientllama-index-embeddings-openaillama-indexmarkitdownmarshmallow-sqlalchemy |
| Maintenance | Actively maintained 92 days since the last release |
| First released | |
| Downloads | 148,733 / month, #11,020 on PyPI 30-day window, as of 2026-08-14 |
| Known vulnerabilities | None known OSV.dev, checked 2026-08-14 |
Evidence: letta-0.16.8-py3-none-any.whl
Tags
Let your AI agent find packages like this
Example. Real query, live index.
You found this page by searching. An agent finds it by wishing: SkillFed indexes 14,416 PyPI packages by what they can do, searchable in plain language.
wish › “llm agents with memory”
- lettaLetta is a framework for building LLM agents with persistent memory…
- memoriMemori is a Python SDK that automatically captures and recalls…
- qwen-agentQwen-Agent is a framework for building LLM applications with tool…
Give your agent the search over MCP, or paste the wish link into any chat.
More Artificial Intelligence packages
LiteLLM provides a unified Python interface to call 100+ LLM providers (OpenAI, Anthropic, Gemini, Bedrock, Azure, and others) using OpenAI-compatible API format, available as both a Python SDK and a self-hosted AI Gateway proxy server.
Install it if you need to work with multiple LLM providers or want to centralize LLM routing in your organization.
Client library and CLI tool for downloading, uploading, and managing models, datasets, and repositories on the Hugging Face Hub platform.
Install it if you work with Hugging Face Hub models or datasets.
LangChain provides a framework for building agents and LLM-powered applications by composing language models, tools, and memory through a unified API that abstracts over multiple model providers.
hf-xet provides chunk-based deduplication and efficient file transfer for the Hugging Face Hub, enabling faster uploads and downloads of large files with local disk caching.
Tokenizers converts raw text into token sequences for NLP models, with support for training custom vocabularies and using pre-built tokenizers (BPE, WordPiece) optimized for speed via Rust.
Transformers provides a unified framework for loading, fine-tuning, and running state-of-the-art pretrained models across text, vision, audio, video, and multimodal tasks using PyTorch, JAX, or TensorFlow.
Install it if you need to run or train any transformer-based model for NLP, vision, audio, or multimodal tasks.
See also letta-client · fast-agent-mcp · memori · PraisonAI · praisonaiagents · langmem · llama-index-agent-openai · deepagents-code · openai-agents · deepagents