---
id: autogen-core
version: "0.7.5"
license: MIT License Copyright (c) Microsoft Corporation. Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to… (full text in the JSON record)
license_treatment: permissive
maintenance: aging
---

# autogen-core — Foundational interfaces and agent runtime implementation for AutoGen

License: permissive · Maintenance: aging · Downloads: 981.4K/mo

## What it is and what it does

AutoGen Core is a foundational runtime library for building multi-agent systems where agents communicate and coordinate through an event-driven, Actor-model architecture. It abstracts the complexity of agent lifecycle management, message passing, and distributed execution, allowing developers to prototype locally and scale to cloud infrastructure without rewriting core agent logic.

The package depends on pydantic for data validation, protobuf for serialization, opentelemetry-api for observability, pillow for image handling, jsonref for schema references, and typing-extensions for modern type hints. It targets developers building conversational AI systems, autonomous workflows, or collaborative agent networks who want a structured, scalable foundation rather than building from scratch.

Use it for:

- Build multi-agent conversational systems where agents coordinate through message passing and event handling.
- Prototype agent workflows locally, then deploy the same code to distributed cloud environments.
- Create resilient agent networks that can recover from failures and scale horizontally.
- Implement complex AI orchestration logic using the Actor model without managing low-level concurrency.
- Monitor and trace agent interactions and performance using OpenTelemetry integration.

## Worth the install?

AI-flagged interpretation of the facts above — verify before relying.

AutoGen Core provides a runtime and foundational interfaces for building event-driven, distributed AI agent systems using the Actor model, enabling local development with a path to cloud deployment.

Yes, if you are building event-driven, distributed agent systems and want a structured Actor-model runtime. The low install friction and permissive MIT license make adoption straightforward. However, the aging maintenance status (318 days since last release) means you should verify that the current version meets your stability and feature needs before committing to production use.

## Install

pip install autogen-core
uv add autogen-core
poetry add autogen-core

## Installing autogen-core

Before you install: Low install friction with a pure-Python wheel. Maintenance status is aging—last release was 318 days ago—so expect slower response to issues, though the package remains functional for current use.

License in practice: MIT License permits free use, modification, and distribution with minimal restrictions, making it suitable for both open-source and commercial projects.

Quickstart:

pip install autogen-core

from autogen_core import Agent
# Define and instantiate an agent using the Actor model runtime

Requires Python 3.10 or later.

Verify before relying:

- Whether the Actor model implementation is production-ready for large-scale distributed deployments.
- Current state of documentation and community examples beyond the official guide.
- Compatibility guarantees or migration path for versions released after 2025-09-30.

## Package facts

- License: MIT License Copyright (c) Microsoft Corporation. Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to… (full text in the JSON record) (permissive)
- Python support: supports_current
- Install friction: low
- Maintenance: aging
- Downloads: 981.4K/month (top 5,000 on PyPI)
- Known vulnerabilities: none known

## Tags

actor model agent framework, distributed ai agent runtime, event-driven agent system, multi-agent orchestration, scalable agent framework, ai agent development platform, agent communication runtime, actor-model, multi-agent-systems, distributed-computing

[View on SkillFed](https://skillfed.io/packages/autogen-core) · [View on PyPI](https://pypi.org/project/autogen-core/)