AutoGen treats a multi agent system as a conversation. Agents message each other and the work gets done through that back and forth. A coding agent writes, an executor runs, a critic reviews, and the loop continues until the group is satisfied.
Reach for it when
- The problem suits debate and critique between agents
- You want code written, executed and reviewed in a loop
- You are on the Microsoft or Azure stack
Look elsewhere when
- You want one deterministic path through a workflow
- Conversation between agents is overhead your task does not need
What the lessons will cover
- 01Install and setup
- 02AssistantAgent and UserProxyAgent
- 03Two agent conversations
- 04Code execution
- 05Group chat and the manager
- 06Tools and function calling
- 07Human in the loop
- 08A working team
Others in multi agent
CrewAIAgents with a role, a goal and a backstory.
MetaGPTAgents assigned software company roles.
CAMEL-AIStudying how large agent societies cooperate.Common questions
Is AutoGen free to use?
AutoGen is open source and free to run yourself. You still pay whichever model provider you point it at, and this tutorial is free with no signup.
Do I need to know Python to use AutoGen?
Basic Python is enough. Functions, dictionaries and imports cover most of what AutoGen asks of you.
When should I not use AutoGen?
You want one deterministic path through a workflow. Conversation between agents is overhead your task does not need.
Lessons for AutoGen are being written. Meanwhile the LangGraph tutorial covers the same ground: state, tools, loops, memory and human approval. Most of it carries straight over.
Start the LangGraph tutorial →