LangChain

title: LangChain subtitle: Using OpenRouter with LangChain headline: ‘LangChain Integration | OpenRouter SDK Support’ canonical-url: https://openrouter.ai/docs/guides/community/langchain og:site_name: OpenRouter Documentation og:title: ‘LangChain Integration - OpenRouter SDK Support’ og:description: ‘Integrate OpenRouter using LangChain framework. Complete guide for LangChain integration with OpenRouter for Python and JavaScript.’ og:image: https://openrouter.ai/dynamic-og?title=LangChain&description=LangChain%20Integration og:image:width: 1200 og:image:height: 630 twitter:card: summary_large_image twitter:site: ‘@OpenRouterAI’ noindex: false nofollow: false

Using LangChain

LangChain provides a standard interface for working with chat models. You can use OpenRouter with LangChain by setting the base_url parameter to point to OpenRouter’s API. For more details on LangChain’s model interface, see the LangChain Models documentation.

Resources:

• Using LangChain for Python: github
• Using LangChain.js: github
• Using Streamlit: github

1
import { ChatOpenAI } from "@langchain/openai";
2
import { HumanMessage, SystemMessage } from "@langchain/core/messages";
3
4
const chat = new ChatOpenAI(
5
  {
6
    model: '<model_name>',
7
    temperature: 0.8,
8
    streaming: true,
9
    apiKey: '${API_KEY_REF}',
10
  },
11
  {
12
    baseURL: 'https://openrouter.ai/api/v1',
13
    defaultHeaders: {
14
      'HTTP-Referer': '<YOUR_SITE_URL>', // Optional. Site URL for rankings on openrouter.ai.
15
      'X-Title': '<YOUR_SITE_NAME>', // Optional. Site title for rankings on openrouter.ai.
16
    },
17
  },
18
);
19
20
// Example usage
21
const response = await chat.invoke([
22
  new SystemMessage("You are a helpful assistant."),
23
  new HumanMessage("Hello, how are you?"),
24
]);