Set Up LangChain & LangGraph with MonoRouter

Route your LangChain agents and LangGraph workflows through your Claude Pro or Max subscription. Two env vars, zero code changes.

What you need

• •A Claude Pro or Max subscription at claude.ai
• •Python 3.9+ with pip
• •A MonoRouter account — sign up at /signup
• •A MonoRouter API key (mrk_*) — get one at /keys
• •Your Claude subscription connected to MonoRouter — go to /tokens, click [ + add token ], and run the auto-connect script

Sign up for MonoRouter

Open /signup, enter your email and set a password. Save the six recovery codes — they're your offline fallback if you lose your password; you can also always reset it by email at /forgot-password.

Connect your Claude subscription

From the dashboard, go to /tokens, click [ + add token ], and pick ANTHROPIC. Use the auto-connect tab — it's the fastest.

The modal shows a one-liner like this:

curl -sSL https://api.monorouter.dev/connect.sh | bash -s -- mrc_xxxxxxxxxxxx

Paste it into a terminal. The script runs claude setup-token, which opens a browser window for authorization and creates a long-lived token. When the dashboard flips to connected, you're done.

Install the packages

pip install langchain-anthropic langgraph

langchain-anthropic is the official LangChain integration for Claude. langgraph is optional — install it if you're building stateful agents with tool use.

Set environment variables

Add these to your shell config or .env file:

export ANTHROPIC_BASE_URL="https://api.monorouter.dev"
export ANTHROPIC_API_KEY="mrk_your_key_here"

Basic LangChain usage

Simple completion through MonoRouter:

from langchain_anthropic import ChatAnthropic

llm = ChatAnthropic(model="claude-sonnet-4-6")

response = llm.invoke("What is MonoRouter?")
print(response.content)

The ChatAnthropic class picks up ANTHROPIC_BASE_URL and ANTHROPIC_API_KEY from the environment. You can also pass them explicitly:

llm = ChatAnthropic(
    model="claude-sonnet-4-6",
    anthropic_api_url="https://api.monorouter.dev",
    anthropic_api_key="mrk_your_key_here",
)

LangGraph agent with tools

Build a tool-calling agent that routes through MonoRouter:

from langchain_anthropic import ChatAnthropic
from langchain_core.tools import tool
from langgraph.prebuilt import create_react_agent

@tool
def get_weather(city: str) -> str:
    """Get the current weather for a city."""
    return f"It's sunny in {city}!"

llm = ChatAnthropic(model="claude-sonnet-4-6")
agent = create_react_agent(llm, tools=[get_weather])

result = agent.invoke({
    "messages": [{"role": "user", "content": "What's the weather in Tokyo?"}]
})

for msg in result["messages"]:
    print(msg.content)

Every LLM call the agent makes routes through MonoRouter automatically. You can see each call on your /dashboard.

Streaming

MonoRouter supports streaming out of the box:

from langchain_anthropic import ChatAnthropic

llm = ChatAnthropic(model="claude-sonnet-4-6", streaming=True)

for chunk in llm.stream("Explain quantum computing in one paragraph"):
    print(chunk.content, end="", flush=True)

Available models

Use any of these model IDs with ChatAnthropic:

Pricing

• •1,500 successful calls are free for the lifetime of your account. Failed calls don't count.
• •After that, paid plans start at $9/month (Hobby, 10,000 calls). Pro is $29/month for unlimited personal use; Team is $99/month for 5 pooled seats. See /billing for the payment flow.