LangChain Integration

Route all LangChain LLM calls through the Curate-Me gateway for automatic cost tracking, rate limiting, and PII scanning. One parameter changed — base_url.

Before and after


# Before — direct to OpenAI
from langchain_openai import ChatOpenAI
 
llm = ChatOpenAI(model="gpt-4o")
 
# After — through Curate-Me gateway
from langchain_openai import ChatOpenAI
 
llm = ChatOpenAI(
    model="gpt-4o",
    base_url="https://api.curate-me.ai/v1/openai",          # ← added
    api_key="sk-your-openai-key",
    default_headers={"X-CM-API-Key": "cm_sk_YOUR_KEY"},      # ← added
)

Every chain, agent, and tool call that uses this LLM instance is now governed by the gateway.

Full example


from langchain_openai import ChatOpenAI
from langchain_core.prompts import ChatPromptTemplate
 
# Point LangChain at the gateway
llm = ChatOpenAI(
    model="gpt-4o",
    base_url="https://api.curate-me.ai/v1/openai",
    api_key="sk-your-openai-key",
    default_headers={
        "X-CM-API-Key": "cm_sk_YOUR_KEY",
        "X-CM-Tags": "project=langchain-app,env=production",  # optional cost tags
    },
    max_tokens=150,
)
 
# Simple invocation
response = llm.invoke("What is an AI gateway proxy?")
print(response.content)
 
# Streaming
for chunk in llm.stream("List three benefits of LLM cost tracking."):
    print(chunk.content, end="", flush=True)
 
# Chains work identically
prompt = ChatPromptTemplate.from_messages([
    ("system", "You are a concise technical writer. Reply in one paragraph."),
    ("human", "{question}"),
])
chain = prompt | llm
result = chain.invoke({"question": "How does rate limiting protect AI apps?"})
print(result.content)

LangGraph agents

LangGraph agents use the same ChatOpenAI instance. Swap the base URL once and every agent step goes through the gateway:


from langgraph.prebuilt import create_react_agent
 
agent = create_react_agent(llm, tools=[...])
result = agent.invoke({"messages": [("user", "Research AI governance trends")]})

TypeScript (LangChain.js)


import { ChatOpenAI } from '@langchain/openai';
 
const llm = new ChatOpenAI({
  model: 'gpt-4o',
  configuration: {
    baseURL: 'https://api.curate-me.ai/v1/openai',
    defaultHeaders: { 'X-CM-API-Key': 'cm_sk_YOUR_KEY' },
  },
  openAIApiKey: 'sk-your-openai-key',
});
 
const response = await llm.invoke('What is AI governance?');
console.log(response.content);

Prerequisites


pip install langchain-openai

What you get

Every LangChain call through the gateway automatically receives:

Feature	Description
Cost tracking	Per-request token and dollar cost recorded in real time
Rate limiting	Per-org, per-key request throttling
PII scanning	Regex scan for secrets and PII before hitting the provider
Model allowlists	Only approved models per your org policy
Budget caps	Daily and monthly spend limits enforced
Audit trail	Full request metadata logged for compliance

Working examples

Python example — non-streaming, streaming, chains
TypeScript example — ChatOpenAI with chains

Next steps

Gateway Quickstart — full setup walkthrough
CrewAI Integration — multi-agent crews with per-agent cost tags
Cost Tracking — budget alerts and cost attribution