AutoGen
AI agent framework for Python with local and distributed runtimes, a browser-based prototyping UI, and OpenAI and Azure OpenAI integrations.
AutoGen is a framework for developers building AI agents that work together or alongside people. Its agent runtime can run locally or across distributed systems, while model integrations such as OpenAI and Azure OpenAI send requests to external services. It's community-managed and in maintenance mode, with no further features or enhancements planned.
The framework lets developers choose how much control they need. AgentChat provides a higher-level Python interface for conversational assistants, two-agent exchanges and group chats. Core handles event-driven workflows and message passing for more customized systems, including agents distributed across Python and .NET applications.
AutoGen Studio provides a browser-based interface for building and running multi-agent prototypes without writing code. It's a prototyping tool rather than a production application. Developers can use AutoGen Bench to evaluate agent performance.
Extensions connect agents to external tools and services. MCP support includes integrations such as Playwright for web browsing, while a Docker code executor runs model-generated code in containers. Developers can also create their own extensions. The documented OpenAI integration supports GPT-4o and requires an OpenAI account and API key, so that setup depends on a cloud service.
AutoGen requires Python 3.10 or later. Code is MIT licensed; documentation and other repository content use Creative Commons Attribution 4.0. Microsoft recommends Microsoft Agent Framework for new projects.
