AutoGen
In plain wordsA free Microsoft software framework for building teams of AI agents that solve problems by messaging one another, writing and running code, and asking people for help when needed.
An open-source Microsoft ResearchMicrosoft Research, prop. n.: Microsoft's research division, with laboratories on several continents, whose scientists have produced influential work on speech, vision and language AI. framework that organizes a multi-agent applicationmulti-agent system, n.: A setup in which several AI agents, each with its own role or point of view, work together — or compete — on a problem, passing messages and results among themselves. as a conversation: language-modellarge language model, n.: A computer program trained on enormous amounts of text to predict the words that come next. At vast scale, that simple skill lets it chat, answer questions, translate, summarize and write code. agents, tool-running agents and people pass messages to one another until the task is done. In the original design, 'conversable agents' — for example an assistant agent that proposes solutions and a user-proxy agent that executes the code it writes and reports the results, optionally consulting a human — converse in two-agent chats or in group chats steered by a manager that picks the next speaker, a pattern that made code-writing and code-executing loops easy to set up. Version 0.4 (January 2025) rebuilt the framework on an asynchronous, event-driven actor model with layered APIs (Core, AgentChat and Extensions) and a no-code AutoGen Studio. In October 2025 Microsoft announced Microsoft Agent Framework, building on AutoGen and its Semantic Kernel SDK, as the successor for new development, and AutoGen moved to community-managed maintenance mode.
In AutoGen we paired a coding agent with an executor agent that ran each script in a Docker sandbox.
Origin
Developed by Chi Wang, Qingyun Wu and colleagues, initially inside Microsoft's FLAML automated machine-learning library, and described in an August 2023 paper; it quickly became one of the most popular multi-agent frameworks. In late 2024 some of its original developers continued the earlier design in a fork named AG2.
Usage note
An example of an agent frameworkagent framework, n.: A software library that gives developers ready-made parts for building AI agents — model connections, tool handling, memory, multi-agent coordination and logging — so they need not write everything from scratch.; compare LangGraphLangGraph, prop. n.: A free software library for building AI agents as flowcharts of steps that share a memory, so developers control exactly how an agent moves between steps, pauses and resumes. (explicit graphs) and CrewAICrewAI, prop. n.: A free software framework for setting up teams of AI agents, each given a job title, goal and backstory, that divide up tasks and work through them together like a small staff. (role-based crews). AutoGen versions 0.2 and 0.4 differ substantially.
Related
- agent frameworkagent framework, n.: A software library that gives developers ready-made parts for building AI agents — model connections, tool handling, memory, multi-agent coordination and logging — so they need not write everything from scratch.
- multi-agent systemmulti-agent system, n.: A setup in which several AI agents, each with its own role or point of view, work together — or compete — on a problem, passing messages and results among themselves.
- LangGraphLangGraph, prop. n.: A free software library for building AI agents as flowcharts of steps that share a memory, so developers control exactly how an agent moves between steps, pauses and resumes.
- CrewAICrewAI, prop. n.: A free software framework for setting up teams of AI agents, each given a job title, goal and backstory, that divide up tasks and work through them together like a small staff.
- orchestrationorchestration, n.: Coordinating several AI calls, tools or agents so each does its part in the right order and their results are combined into one finished piece of work, as a conductor directs an orchestra.
- AI agentAI agent, n.: An AI system that does not just answer but acts: given a goal, it decides on steps, uses tools such as a browser or code editor, checks the results and keeps going until the job is done.
Sources
- Wu, Q. et al. (2023). AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversation. arXiv:2308.08155.
Mentioned in
CrewAICrewAI, prop. n.: A free software framework for setting up teams of AI agents, each given a job title, goal and backstory, that divide up tasks and work through them together like a small staff., LangGraphLangGraph, prop. n.: A free software library for building AI agents as flowcharts of steps that share a memory, so developers control exactly how an agent moves between steps, pauses and resumes.