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

# Quickstart

> Get started with Auto in minutes

## Step 1: Get Your API Key

You can get your Auto API key from [your dashboard](https://auto.venki.dev/runs).

## Step 2: Proxy your requests through Auto

You'll need to make some simple changes to your LLM call to integrate with Auto.

<CodeGroup>
  ```typescript OpenAI SDK (.ts) theme={"system"}
  const openai = new OpenAI({
    // #1: Switch to the Auto API key
    apiKey: process.env.AUTO_API_KEY, 
    // #2: Add the Auto base URL
    baseURL: "https://api.auto.venki.dev/api/v1", 
    // #3: Add the OpenAI API key - optional, see Bringing Your Own Keys for more details
    defaultHeaders: {
      "X-OpenAI-Api-Key": process.env.OPENAI_API_KEY, 
    },
  });

  const chatResponse = await openai.chat.completions.create(
    {
      model: "gpt-4o"
      messages: ...,
    },
    {
      // #4: Add a prompt ID
      headers: {
        "X-Auto-Prompt-Id": "summarize-webpage", 
      },
    }
  );
  ```

  ```typescript Fetch API (.ts) theme={"system"}
  // #1: Switch to the Auto base URL
  const response = await fetch("https://api.auto.venki.dev/api/v1/chat/completions", {
    method: "POST",
    headers: {
      // #2: Switch to the Auto API key
      "Authorization": `Bearer ${process.env.AUTO_API_KEY}`,
      "Content-Type": "application/json",
      // #3: Add the OpenAI API key - optional, see Bringing Your Own Keys for more details
      "X-OpenAI-Api-Key": process.env.OPENAI_API_KEY,
      // #4: Add a prompt ID 
      "X-Auto-Prompt-Id": "summarize-webpage"
    },
    body: JSON.stringify({
      model: "gpt-4o",
      messages: ...,
    })
  });

  const chatResponse = await response.json();
  ```

  ```typescript Vercel AI SDK (.ts) theme={"system"}
  const openai = createOpenAI({
    // #1: Switch to the Auto API key
    apiKey: process.env.AUTO_API_KEY,
    // #2: Add the Auto base URL
    baseURL: "https://api.auto.venki.dev/api/v1",
    // #3: Add the OpenAI API key - optional, see Bringing Your Own Keys for more details
    headers: {
      "X-OpenAI-Api-Key": process.env.OPENAI_API_KEY,
    },
  });

  const { text } = await generateText({
    model: openai('gpt-4o'),
    prompt: '...',
    headers: {
      "X-Auto-Prompt-Id": "summarize-webpage"
    }
  });

  ```
</CodeGroup>

### Choosing a Prompt ID

Auto aggregates results at the prompt ID level, and picks the best model for each prompt ID. We thus recommend that you pick a unique, descriptive string for each distinct use case in your codebase.

Generally, if you're unsure about whether to reuse the prompt ID for different use cases in your codebase, it's better to err towards creating a new prompt ID.

## Step 3: You're All Set! 🎉

You should start seeing logs immediately, and leaderboards in a few days.