ag2
AG2 is an open-source AI agent operating system that supports multi-agent collaboration and various large language models.
- Type
- MCP
- Open source
- Yes
- GitHub Stars
- ★ 4.7k
- Source
- mcp-github
- Repository
- github.com/ag2ai/ag2
Overview
AG2 (formerly AutoGen) is an open-source programming framework for building AI agents and enabling collaboration among multiple agents to accomplish tasks. It provides features such as agent interaction, support for various large language models (LLM), tool integration, workflows with both autonomous and human involvement, and multi-agent conversation patterns. AG2 aims to simplify the development and research of agent-based AI. Users can get started by installing the Python package and configuring API keys. Ideal for developers who need to build complex AI agent systems.
Capabilities
- ▪Multi-agent collaboration
- ▪Support for multiple large language models
- ▪Tool usage support
- ▪Workflows with autonomous and human participation
- ▪Multi-agent conversation patterns
Use cases
Setup
pip install ag2[openai]
This information was compiled by AI from public sources and may contain inaccuracies — please refer to the source.
FAQ
Which large language models does AG2 support?
AG2 supports multiple large language models, such as GPT-5.
How do I configure API keys?
Store your API keys in a configuration file, for example, OAI_CONFIG_LIST.
Related skills
gemini-cli
Gemini CLI is an open-source AI agent that brings Gemini's powerful capabilities directly into the terminal.
private-gpt
Provides a complete API layer for local models, supporting RAG, skills, tools, and more.
planning-with-files
File-based persistent planning skill, ideal for AI coding agents and long-running tasks.
scientific-agent-skills
Transform any AI agent into a scientific assistant with 147 ready-to-use research skills.
susi_alexa_skill
A skill that enables question-and-answer interactions between Alexa and Susi AI.
claude-context
Provides code search capabilities for Claude Code, turning the entire codebase into context.