AutoGen by Microsoft
Microsoft's open-source multi-agent framework
Overview
AutoGen is Microsoft Research's framework for building multi-agent LLM applications. Strong on conversational agent patterns, agent-to-agent negotiation, code-execution workflows. Open-source, with growing adoption inside Microsoft's own products.
Pricing
Pricing shown for reference only. These figures reflect RECATOOLS research as of 19 May 2026 and may be out of date or incomplete. This is not financial or purchasing advice — always confirm the current price on the provider’s official website before making any decision.
Use cases
What you can produce with AutoGen by Microsoft
- Define two or more LLM agents in Python — for example a coder and a reviewer — and have them converse autonomously until a task is solved.
- Build a group chat where a manager agent routes a problem among specialist agents (researcher, writer, critic) and returns the consolidated answer.
- Let an agent write Python code and execute it in a sandboxed Docker container, feeding errors back for automatic self-correction.
- Insert a human-in-the-loop checkpoint so an agent pauses for your approval before executing code or committing to an answer.
- Prototype and test multi-agent workflows visually in AutoGen Studio's low-code interface without writing orchestration code.
- Attach custom tools and function calls to agents so they can query APIs, read files or hit databases during a conversation.
- Study the extensive open-source examples to learn multi-agent design patterns you can carry into Microsoft Agent Framework, its production successor.
ASEAN Perspective
AutoGen by Microsoft in Southeast Asia
ASEAN-region availability and pricing notes coming soon. Drop the editorial team a note via /contact/ if you can supply local context (Singapore/Malaysia/Indonesia/Thailand/Vietnam).
AutoGen is Microsoft's open-source framework for building multi-agent LLM applications, enabling agents that converse, use tools, and collaborate to solve tasks, with an event-driven core, AgentChat layer, and Studio UI. It is a leading research-grade toolkit for orchestrating multiple cooperating agents and is free.
It suits AI engineers and researchers prototyping agentic systems who are comfortable with Python and an evolving API; it is a developer framework, not a finished product, and the rapid version churn means breaking changes and a real learning curve. Production hardening is on you. Open-source and self-hostable, so ASEAN use and data control are unrestricted; you pay only for the underlying model calls.
What people say
AutoGen deserves real credit as the framework that popularised multi-agent LLM development — its conversational agent pattern, group-chat orchestration and sandboxed code execution shaped how the whole industry thinks about agent teams. But its status changed decisively: in October 2025 Microsoft announced the Microsoft Agent Framework, a public-preview convergence of AutoGen's orchestration ideas with Semantic Kernel's enterprise foundations, and shipped version 1.0 for Python and .NET in April 2026. AutoGen itself is now in maintenance mode, with all new development flowing into the unified framework.
Developer sentiment over AutoGen's lifetime was a mix of admiration and frustration. Practitioners praised how quickly it let them prototype agent societies — two agents debating, a coder agent writing and executing Python, a critic agent reviewing output — often in a few dozen lines. AutoGen Studio's low-code interface lowered the barrier further. The loudest complaint was churn: the v0.2 to v0.4 transition was a ground-up rewrite with breaking changes, and GitHub discussions from that period are full of migration pain, missing v0.2 features such as the local command-line code executor, and documentation that lagged the moving API. Teams that built on v0.2 effectively had to port twice — once to v0.4, and now conceptually to Agent Framework.
As open-source software, AutoGen remains free, functional and installable, and the enormous body of tutorials and examples still has pedagogical value. Microsoft's own migration guidance is explicit that new production work should target the Microsoft Agent Framework instead.
AutoGen fits researchers and learners studying multi-agent patterns, and teams maintaining existing AutoGen deployments that are not ready to migrate. Anyone starting a new agent project in the Microsoft ecosystem should begin with Agent Framework directly; anyone allergic to framework churn may prefer LangGraph or CrewAI, though those evolve quickly too.
Summary of public user & expert reviews, compiled by RECATOOLS.
About this listing
This entry was compiled from publicly available data including AutoGen by Microsoft's official website, press releases, documentation, and reputable third-party publications. RECATOOLS is not affiliated with AutoGen by Microsoft unless explicitly stated.
Third-party AI tools update their pricing, features, availability, and policies frequently. Information here may be outdated by the time you read this — we make reasonable efforts to keep listings current, but cannot guarantee absolute accuracy.
For the latest details, please refer to AutoGen by Microsoft directly →
Spotted something out of date? Suggest an update →
AutoGen by Microsoft in the news
Cybersecurity
Microsoft's Largest-Ever Patch Tuesday: 600-Plus Fixes, Two Live Zero-Days, and a Triage P...
Cybersecurity
SharePoint RCE CVE-2026-45659 Added to CISA KEV: A 'Less Likely' May Patch Is Now Actively...
Developer Tools
Miasma: How a Hijacked CI Pipeline Shipped Malicious npm Packages With Valid Provenance
Alternatives to AutoGen by Microsoft
More in Agents & Automation
