AG2 (formerly AutoGen)¶

What it is¶

AG2 is the next-generation evolution of the AutoGen framework. It is an open-source framework for building multi-agent AI applications that can converse with each other and interact with tools and environments. As of July 2026, it serves as a universal runtime (AG2 AgentOS) for orchestrating specialized agents from various frameworks, fully integrated with Gemma 3 for local reasoning and the MCP 3.0 Task Protocol.

What problem it solves¶

It simplifies the development of complex AI systems where multiple agents need to collaborate, reason, and act. AG2 addresses "islands of intelligence" by providing a universal runtime for framework interoperability, unified state management ("shared brain"), and standardized protocols (A2A and MCP) for secure agent-to-agent and agent-to-tool communication. It specifically solves the orchestration bottleneck in large-scale agentic deployments.

Where it fits in the stack¶

Framework / Multi-Agent Orchestrator / Agent Runtime.

Typical use cases¶

Multi-Framework Orchestration: Connecting agents built in different frameworks (e.g., a LangChain researcher and an OpenAI analyst) into a single cohesive team.
Cross-Platform Coordination: Assembling dynamic teams of specialized personas that can operate across local (Gemma 3) and cloud environments.
Unified State Management: Maintaining consistent context and task state across long-running agentic workflows.
Visual Team Composition: Using Waldiez (the community-led visual companion) to design and debug multi-agent group chats.

Strengths¶

Protocol-First Interoperability: Native support for A2A (Agent-to-Agent) and MCP 3.0 Task Protocol.
Flexible Conversational Design: Support for group chats, hierarchical orchestration, and custom state-based transitions.
Enterprise-Ready Security: Features like Agent Cards and secure tool-calling guards for production deployments.
Shared Brain Architecture: Advanced state management that prevents context loss in complex multi-step tasks.

Limitations¶

Transition Complexity: Migrating from legacy AutoGen (v0.2) to the AG2 AgentOS architecture requires refactoring of orchestration logic.
Orchestration Overhead: The high level of abstraction can make fine-grained control over individual LLM parameters more complex than using low-level SDKs.

When to use it¶

When you need to build sophisticated multi-agent systems involving agents from multiple different providers or frameworks.
When you require a proven, enterprise-grade foundation for collaborative AI workflows.
When building AI-native organizations where specialized agents must discover and delegate to each other dynamically.

When not to use it¶

For simple, single-agent tasks where a direct SDK call is sufficient.
If you prefer a rigid DAG-based workflow without conversational flexibility.

Getting started¶

Installation¶

pip install ag2

Basic Multi-Agent Setup¶

AG2 maintains compatibility with the autogen package name:

import autogen
from ag2 import AgentOS

# Initialize the universal runtime
runtime = AgentOS.init()

# Define agents
assistant = autogen.AssistantAgent("helper", llm_config={"model": "gpt-4o"})
user_proxy = autogen.UserProxyAgent("user", code_execution_config={"use_docker": False})

# Orchestrate
user_proxy.initiate_chat(assistant, message="Analyze our cross-framework dependencies.")

CLI examples¶

Initializing a Project¶

ag2 init my-agent-org

Running in Studio Mode¶

ag2 studio --port 8081

Managing Agent Cards¶

ag2 cards list

API examples¶

Cross-Framework Delegation (A2A)¶

from ag2.protocols import A2A

# Delegate a task to an external Google ADK agent
result = await A2A.delegate(
    target_agent_id="google-adk-analyst",
    task="Perform financial sentiment analysis",
    context=shared_brain.get_context()
)

Using Gemma 3 for Local Reasoning¶

import autogen
from ag2 import LocalRuntime

# Setup local Gemma 3 agent via MCP
runtime = LocalRuntime.use_model("gemma-3-27b")
agent = autogen.AssistantAgent("local-reasoner", llm_config=runtime.config)

Unified State Access¶

# Access the 'shared brain' across the team
state = runtime.get_state("workflow-id-123")
print(state.history)

Gemma 3 — Canonical local LLM for agentic reasoning.
AutoGen — The original legacy framework.
CrewAI — Role-based multi-agent framework.
LangGraph — Graph-based agent orchestration.
Mastra — TypeScript-native agent framework.
Rivet — Visual AI programming environment.
MCP — Standardized tool-calling protocol.
PydanticAI — Type-safe agent framework.
Semantic Kernel — Microsoft's agentic framework.

Sources / references¶

Contribution Metadata¶

Last reviewed: 2026-07-21
Confidence: high