LangGraph

Build stateful, graph-based agents with LangGraph and LangChain.

Language:PythonProviders:OpenAI, Anthropic, Google

Quick start

1

Create the project

npx agentvoy create my-project --framework langgraph --provider openai --model gpt-4o --yes
2

Install and run

cd my-project-agent
pip install -r requirements.txt
cp .env.example .env
# Add your API key
python run.py

What gets generated

LangGraph agents use a StateGraph with tool-calling nodes:

from langgraph.graph import StateGraph, END
from langchain_openai import ChatOpenAI

llm = ChatOpenAI(model=os.environ.get("DEFAULT_MODEL", "gpt-4o"))

def create_graph():
    graph = StateGraph(AgentState)
    graph.add_node("agent", agent_node)
    graph.add_node("tools", tool_node)
    graph.set_entry_point("agent")
    graph.add_conditional_edges("agent", should_continue, {
        "tools": "tools",
        "end": END,
    })
    graph.add_edge("tools", "agent")
    return graph.compile()

Provider support

ProviderLangChain classGuide
OpenAIChatOpenAIGet key
AnthropicChatAnthropicGet key
GoogleChatGoogleGenerativeAIGet key

When to use LangGraph

  • You need complex, multi-step agent workflows with branching logic
  • You want fine-grained control over the agent's decision graph
  • You're already using LangChain and want to add agent capabilities