AI AgentTools & ProtocolsUpdated 2026.04.28

AutoGen

Also known asMicrosoft AutoGen

In one line

AutoGen is Microsoft Research's multi-agent framework, built around conversational agents that exchange messages and divide work between themselves.

Going deeper

AutoGen is Microsoft Research's multi-agent framework, first released in 2023. Its core abstraction is the 'conversable agent' — every action is reduced to messages exchanged between agents, which makes collaboration, review and code execution emerge naturally from conversation rather than from special-cased flow control.

The pattern AutoGen is best known for is UserProxy plus AssistantAgent: a stand-in for the human and an LLM assistant trade messages, run code and verify results until the task is done. It works well for code generation and data analysis flows where iteration and execution are tightly coupled.

AutoGen 0.4 reorganised the architecture heavily and added production-leaning features like distributed execution and event-driven messaging. Where CrewAI is anchored on role definitions, AutoGen is anchored on the message protocol — that is a fair shorthand for the difference.

Related terms

How does your brand show up in AI answers?

Villion measures how your brand appears across ChatGPT, Perplexity and AI Overviews, then automates the work that lifts citation rate and share of voice.

Get a free audit