> ## 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.

# Anthropic SDK

> Route your Anthropic SDK requests through LLM Router to add context compression, PII redaction, and fallback capabilities to Claude.

***

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

If your codebase is already heavily integrated with the official `@anthropic-ai/sdk`, you don't need to rewrite it to use the OpenAI format. LLM Router provides a fully compatible Anthropic endpoint.

By routing the Anthropic SDK through LLM Router, you instantly gain **PII Redaction**, **Context Compression**, and **Automatic Fallbacks** (e.g., falling back to Amazon Bedrock if Anthropic's direct API is down).

## Configuration

To use LLM Router, you simply override the `baseURL` when initializing the Anthropic client.

```typescript Node.js theme={null}
import Anthropic from "@anthropic-ai/sdk";

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

async function main() {
  const msg = await anthropic.messages.create({
    // 2. Use standard Anthropic models
    model: "anthropic/claude-3-5-sonnet",
    max_tokens: 1024,
    messages: [{ role: "user", content: "Write a Python script." }],
  });

  console.log(msg.content);
}
main();
```

```python Python theme={null}
import os
from anthropic import Anthropic

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

# 2. Use standard Anthropic models
message = client.messages.create(
    model="claude-3-5-sonnet-20241022",
    max_tokens=1024,
    messages=[{"role": "user", "content": "Write a Python script."}]
)

print(message.content)
```


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.