> ## Documentation Index
> Fetch the complete documentation index at: https://docs.llmrouter.app/llms.txt
> Use this file to discover all available pages before exploring further.

# OpenAI

> Route your OpenAI SDK requests through LLM Router to access any model (Claude, Gemini, Llama) using standard OpenAI formatting.

***

> ## Documentation Index
>
> Fetch the complete documentation index at: [https://docs.llmrouter.app/llms.txt](https://docs.llmrouter.app/llms.txt)

LLM Router is 100% compatible with the official OpenAI SDK. This means you can use the familiar `openai` package in Node.js or Python to access **any** model (including Anthropic's Claude and Google's Gemini) without changing your code structure.

## 1. Chat Completions API

The Chat Completions API is the standard way to interact with LLMs. To use LLM Router, you simply change the `baseURL` and `apiKey` when initializing the client.

<CodeGroup>
  ```typescript Node.js theme={null}
  import OpenAI from "openai";

  // 1. Initialize with LLM Router credentials
  const client = new OpenAI({
    baseURL: "https://api.llmrouter.app/v1",
    apiKey: process.env.LLM_ROUTER_API_KEY,
  });

  async function main() {
    const response = await client.chat.completions.create({
      // 2. Use ANY supported model (prefixed with provider slug)
      model: "anthropic/claude-3-5-sonnet",
      messages: [{ role: "user", content: "Explain quantum physics." }],

      // 3. (Optional) Pass LLM Router features natively at the root
      // @ts-expect-error - Custom LLM Router extension
      gateway: {
        chatHistoryCompression: { enabled: true, score: 0.6 },
        redact: { token: true },
      },
    });

    console.log(response.choices[0].message.content);
  }
  main();
  ```

  ```python Python theme={null}
  from openai import OpenAI
  import os

  # 1. Initialize with LLM Router credentials
  client = OpenAI(
      base_url="https://api.llmrouter.app/v1",
      api_key=os.environ.get("LLM_ROUTER_API_KEY")
  )

  # 2. Use ANY supported model (prefixed with provider slug)
  response = client.chat.completions.create(
      model="anthropic/claude-3-5-sonnet",
      messages=[{"role": "user", "content": "Explain quantum physics."}],

      # 3. (Optional) Pass LLM Router features natively
      gateway={
          "chatHistoryCompression": { "enabled": True, "score": 0.6 },
          "redact": { "token": True }
      }
  )

  print(response.choices[0].message.content)
  ```
</CodeGroup>

***

## 2. OpenAI Messages API

If you are using the new OpenAI Developer Messages API (often used for specific agentic workflows or structured outputs), LLM Router supports this endpoint natively as well.

You use the exact same client configuration, but call the `client.beta.messages` methods.

<CodeGroup>
  ```typescript Node.js theme={null}
  import OpenAI from "openai";

  const client = new OpenAI({
    baseURL: "https://api.llmrouter.app/v1",
    apiKey: process.env.LLM_ROUTER_API_KEY,
  });

  async function main() {
    const response = await client.beta.messages.create({
      model: "openai/o1-mini",
      messages: [{ role: "user", content: "Write a sorting algorithm." }],
      max_tokens: 1024,

      // Send routing logic and context compression rules
      // @ts-expect-error - Custom LLM Router extension
      gateway: {
        chatHistoryCompression: { enabled: true, score: 0.5 },
      },
    });

    console.log(response.content[0].text);
  }
  main();
  ```

  ```python Python theme={null}
  from openai import OpenAI
  import os

  client = OpenAI(
      base_url="https://api.llmrouter.app/v1",
      api_key=os.environ.get("LLM_ROUTER_API_KEY")
  )

  response = client.beta.messages.create(
      model="openai/o1-mini",
      max_tokens=1024,
      messages=[{"role": "user", "content": "Write a sorting algorithm."}],

      # Custom LLM Router configuration
      gateway={
          "chatHistoryCompression": { "enabled": True, "score": 0.5 }
      }
  )

  print(response.content[0].text)
  ```
</CodeGroup>

***