AutoGen

Skip to main content Back to top Ctrl+K GitHub Discord Twitter AutoGen# AutoGen A framework for building AI agents and applications Studio An web-based UI for prototyping with agents without writing code. Built on AgentChat. pip install -U autogenstudio autogenstudio ui --port 8080 --appdir ./myapp Start here if you are new to AutoGen and want to prototype with agents without writing code. Get Started AgentChat A programming framework for building conversational single and multi-agent applications. Built on Core. Requires Python 3.10+. # pip install -U "autogen-agentchat" "autogen-ext[openai]" import asyncio from autogen_agentchat.agents import AssistantAgent from autogen_ext.models.openai import OpenAIChatCompletionClient async def main() -> None: agent = AssistantAgent("assistant", OpenAIChatCompletionClient(model="gpt-4o")) print(await agent.run(task="Say 'Hello World!'")) asyncio.run(main()) Start here if you are prototyping with agents using Python. Migrating from AutoGen 0.2?. Get Started Core An event-driven programming framework for building scalable multi-agent AI systems. Example scenarios: Deterministic and dynamic agentic workflows for business processes. Research on multi-agent collaboration. Distributed agents for multi-language applications. Start here if you are getting serious about building multi-agent systems. Get Started Extensions Implementations of Core and AgentChat components that interface with external services or other libraries. You can find and use community extensions or create your own. Examples of built-in extensions: McpWorkbench for using Model-Context Protocol (MCP) servers. OpenAIAssistantAgent for using Assistant API. DockerCommandLineCodeExecutor for running model-generated code in a Docker container. GrpcWorkerAgentRuntime for distributed agents. Discover Community Extensions Create New Extension --- Skip to main content Back to top Ctrl+K GitHub Discord Twitter AgentChat# AgentChat is a high-level API for building multi-agent applications. It is built on top of the autogen-core package. For beginner users, AgentChat is the recommended starting point. For advanced users, autogen-core’s event-driven programming model provides more flexibility and control over the underlying components. AgentChat provides intuitive defaults, such as Agents with preset behaviors and Teams with predefined multi-agent design patterns. Installation How to install AgentChat Installation: How to install AgentChat Quickstart Build your first agent Quickstart: Build your first agent Tutorial Step-by-step guide to using AgentChat, learn about agents, teams, and more Tutorial: Step-by-step guide to using AgentChat, learn about agents, teams, and more Custom Agents Create your own agents with custom behaviors Custom Agents: Create your own agents with custom behaviors Selector Group Chat Multi-agent coordination through a shared context and centralized, customizable selector Selector Group Chat: Multi-agent coordination through a shared context and centralized, customizable selector Swarm Multi-agent coordination through a shared context and localized, tool-based selector Swarm: Multi-agent coordination through a shared context and localized, tool-based selector Magentic-One Get started with Magentic-One Magentic-One: Get started with Magentic-One GraphFlow (Workflow) Multi-agent workflows through a directed graph of agents. GraphFlow: Multi-agent workflows through a directed graph of agents. Memory Add memory capabilities to your agents Memory: Add memory capabilities to your agents Logging Log traces and internal messages Logging: Log traces and internal messages Serialize Components Serialize and deserialize components Serialize Components: Serialize and deserialize components Examples Sample code and use cases Examples: Sample code and use cases Migration Guide How to migrate from AutoGen 0.2.x to 0.4.x. Migration Guide: How to migrate from AutoGen 0.2.x to 0.4.x. Edit on GitHub Show Source --- Skip to main content Back to top Ctrl+K This is documentation for version stable.Switch to stable version GitHub Discord Twitter Core# AutoGen core offers an easy way to quickly build event-driven, distributed, scalable, resilient AI agent systems. Agents are developed by using the Actor model. You can build and run your agent system locally and easily move to a distributed system in the cloud when you are ready. Key features of AutoGen core include: Asynchronous Messaging Agents communicate through asynchronous messages, enabling event-driven and request/response communication models. Scalable & Distributed Enable complex scenarios with networks of agents across organizational boundaries. Multi-Language Support Python & Dotnet interoperating agents today, with more languages coming soon. Modular & Extensible Highly customizable with features like custom agents, memory as a service, tools registry, and model library. Observable & Debuggable Easily trace and debug your agent systems. Event-Driven Architecture Build event-driven, distributed, scalable, and resilient AI agent systems. Edit on GitHub Show Source --- Skip to main content Back to top Ctrl+K This is documentation for version stable.Switch to stable version GitHub Discord Twitter Extensions# AutoGen is designed to be extensible. The autogen-ext package contains the built-in component implementations maintained by the AutoGen project. Examples of components include: autogen_ext.agents.* for agent implementations like MultimodalWebSurfer autogen_ext.models.* for model clients like OpenAIChatCompletionClient and SKChatCompletionAdapter for connecting to hosted and local models. autogen_ext.tools.* for tools like GraphRAG LocalSearchTool and mcp_server_tools(). autogen_ext.executors.* for executors like DockerCommandLineCodeExecutor and ACADynamicSessionsCodeExecutor autogen_ext.runtimes.* for agent runtimes like GrpcWorkerAgentRuntime See API Reference for the full list of components and their APIs. We strongly encourage developers to build their own components and publish them as part of the ecosytem. Discover Discover community extensions and samples Discover: Discover community extensions and samples Create your own Create your own extension Create your own: Create your own extension Edit on GitHub Show Source