Term
AutoGen
AutoGen is an open-source framework from Microsoft for building multi-agent applications with LLMs. Several AI agents solve tasks together in structured conversations — for example group chats or asynchronous message passing.
AutoGen — explained in detail
AutoGen is an open-source framework developed by Microsoft Research that lets multiple LLM-powered agents collaborate in coordinated conversations. Instead of sending a single prompt to a model, you define specialized agents — for example a planner, a coder and a critic — that solve a task iteratively by exchanging messages. With the architecture introduced in version 0.4, AutoGen is layered: an asynchronous, event-driven Core API (an actor model with local and distributed runtime), the simpler, more opinionated AgentChat API for rapid prototyping with two-agent and group chats, and an Extensions API for LLM clients and capabilities such as code execution. As of 2026 the stable line sits at the 0.7.x releases.
An important point for context: according to Microsoft, AutoGen is now in maintenance mode. For new projects Microsoft points to the Microsoft Agent Framework, which converges AutoGen and Semantic Kernel and offers long-term support; an official migration guide exists. Existing AutoGen applications can continue to run.
Example / Practical use
A classic AutoGen pattern is the pair of an AssistantAgent (proposes solutions, writes code) and a UserProxyAgent (runs the code, returns results or error messages). The two exchange messages until the task is done — for instance a data analysis: the assistant suggests a Python script, the proxy runs it, returns a traceback, the assistant fixes it. Group chats let you orchestrate several roles, governed by a manager that decides which agent speaks next.
Distinction from related terms
LangChain is broader in scope (chains, tools, retrieval) and not primarily a multi-agent framework; LangGraph models agent workflows explicitly as graphs with states and edges, whereas AutoGen emphasizes the conversational, message-passing approach. LlamaIndex sits alongside AutoGen with a focus on data connectivity and RAG rather than competing with it. CrewAI follows a similar role-based model but is a separate project outside Microsoft.
Entdecke mehr
Saving Tokens with Claude: 6 Principles That Make Experts Twice as Fast
How I turned my CLAUDE.md from a style guide into a token budget — 6 principles for lower cost, less waiting, and more honest reporting.
GlossarCrewAI
CrewAI is an open-source framework for orchestrating multiple autonomous AI agents. It assigns agents roles, goals and tools and has them collaborate in teams ("crews") to solve complex tasks through coordinated, multi-step workflows.
LexikonLangGraph
LangGraph as the standard for agent orchestration — nodes, edges, state, loops, human-in-the-loop and persistence explained clearly.