AutoGen

Microsoft open-source framework for multi-agent chat apps and event-driven agent systems.

Visit site

Pricing

Open-source framework under MIT; you bring your own model API keys. AutoGen Studio UI available via pip.

Overview

AutoGen is a Microsoft open-source programming framework for building agentic AI applications. AgentChat covers conversational single- and multi-agent apps in Python. Core adds an event-driven runtime for scalable multi-agent systems. Extensions connect MCP servers, OpenAI Assistants, Docker code execution, and distributed workers. AutoGen Studio offers a web UI to prototype agents without writing code.

Key features

  • AgentChat API for conversational multi-agent apps
  • Event-driven Core runtime for scalable systems
  • AutoGen Studio no-code prototyping UI
  • MCP workbench and OpenAI Assistant extensions
  • Docker-based code execution tools
  • Distributed agent runtime options
  • Model-client integrations (e.g. OpenAI)

Use cases

Show all 8Show fewer

Alternatives

  • AgentForgeOpen-source low-code Python framework for agents and YAML cognitive architectures.ai agent frameworkFree
  • IXOpen-source platform for building, debugging, and deploying collaborative GPT agents and workflows.ai agent frameworkFree
  • OpenAI Agents SDKOpen-source SDK for building agents with tools, handoffs, guardrails, and tracing.ai agent framework
  • AgentOpsDeveloper observability platform to trace, debug, and cost-track ai agents.AI toolFreemium
  • AgnoSelf-hosted Python agent platform with AgentOS runtime and browser control plane.ai agent frameworkFreemium
  • ArizeObservability and evaluation platform for tracing, testing, and improving ai agents.AI toolFreemium

Last updated 2026-09-16