How to beginner · 3 min read

How to install LangGraph

Q: How to install LangGraph

Install langgraph via pip install langgraph to use stateful graph-based AI agents in Python. This package enables building and running AI workflows with nodes and edges using a simple API.

Quick answer

Install langgraph via pip install langgraph to use stateful graph-based AI agents in Python. This package enables building and running AI workflows with nodes and edges using a simple API.

PREREQUISITES

Python 3.8+
pip installed

Setup

Install langgraph using pip. Ensure you have Python 3.8 or higher installed. No API key is required for the package itself.

bash

pip install langgraph

output

Collecting langgraph
  Downloading langgraph-<version>-py3-none-any.whl (xx kB)
Installing collected packages: langgraph
Successfully installed langgraph-<version>

Step by step

Here is a complete example to create a simple LangGraph stateful graph with one node that appends a response to messages, then runs the graph.

python

from langgraph.graph import StateGraph, END
from typing import TypedDict

class State(TypedDict):
    messages: list

def my_node(state: State) -> State:
    return {"messages": state["messages"] + ["response"]}

graph = StateGraph(State)
graph.add_node("my_node", my_node)
graph.set_entry_point("my_node")
graph.add_edge("my_node", END)

app = graph.compile()
result = app.invoke({"messages": ["Hello"]})
print(result)

output

{'messages': ['Hello', 'response']}

Common variations

You can define more complex state types with TypedDict and add multiple nodes and edges to build complex workflows. The invoke method runs the compiled graph synchronously. For persistence, use a checkpointer like MemorySaver when compiling.

python

from langgraph.graph import StateGraph, END
from langgraph.checkpointers.memory import MemorySaver
from typing import TypedDict

class State(TypedDict):
    count: int

def increment_node(state: State) -> State:
    return {"count": state["count"] + 1}

graph = StateGraph(State)
graph.add_node("increment", increment_node)
graph.set_entry_point("increment")
graph.add_edge("increment", END)

checkpointer = MemorySaver()
app = graph.compile(checkpointer=checkpointer)

result = app.invoke({"count": 0})
print(result)

output

{'count': 1}

Troubleshooting

If you see ModuleNotFoundError: No module named 'langgraph', ensure you installed the package with pip install langgraph in the correct Python environment.
If your graph does not run, verify that you set an entry point with set_entry_point() and added edges to END.
For type errors, confirm your TypedDict state matches the node function input and output.

✅

Key Takeaways

Install LangGraph with pip install langgraph for stateful AI agent graphs.
Define your state with TypedDict and add nodes and edges to build workflows.
Use graph.compile() and app.invoke() to run your graph synchronously.
Set an entry point and edges to END to ensure graph execution completes.
Troubleshoot missing modules by verifying your Python environment and installation.

Verified 2026-04

Verify ↗