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

# Use the LLM Gateway in code

> Integrate the LLM Gateway into your applications using OpenAI-compatible SDKs

Set up your application to interact with an LLM.

<Steps>
  <Step title="Set up your account">
    Vist your <a href="/developers/build-agents/create-account" target="_blank">ATXP account dashboard</a> and copy your account connection string into an environment variable. The best way to do this is to create a `.env` file in the root of your project and add the following line:

    ```bash .env lines theme={null}
    ATXP_CONNECTION=https://accounts.atxp.ai?connection_token=<random_string>&account_id=<random_string>
    ```

    <Warning>
      Never commit your `.env` file to version control. It is a good idea to add your `.env` to your `.gitignore` file to prevent it from being committed.

      ```bash theme={null}
      echo .env >> .gitignore
      ```
    </Warning>
  </Step>

  <Step title="Use the gateway">
    <Tabs>
      <Tab title="JS OpenAI SDK">
        Install the OpenAI SDK in your project.

        ```bash theme={null}
        npm install openai
        ```

        Create a client using an OpenAI's SDK and use it.

        ```typescript theme={null}
        // Import the OpenAI SDK
        const { OpenAI } = await import('openai');

        // Create the client
        const openai = new OpenAI({
          apiKey: process.env.ATXP_CONNECTION,
          baseURL: 'https://llm.atxp.ai/v1'
        });

        // Chat with the LLM
        const result = await openai.chat.completions.create({
          model: 'gpt-4.1',
          messages: [
            { role: 'system', content: 'You are a helpful assistant.' },
            { role: 'user', content: 'What is 2+2?' }
          ],
        });

        // Process the result
        console.log(result.choices[0].message.content);
        ```
      </Tab>

      <Tab title="JS Vercel SDK">
        Install the Vercel AI SDK and openai-compatible package in your project.

        ```bash theme={null}
        npm install ai @ai-sdk/openai-compatible
        ```

        Create an openai-compatible client using an Vercel's SDK and use it.

        ```typescript theme={null}
        // Import Vercel's SDK and openai-compatible adapter
        const { createOpenAICompatible } = await import('@ai-sdk/openai-compatible');
        const { generateText } = await import('ai');

        // Create the client
        const atxp = createOpenAICompatible({
          name: 'atxp-llm',
          apiKey: process.env.ATXP_CONNECTION,
          baseURL: 'https://llm.atxp.ai/v1'
        });

        // Chat with the LLM
        const { text } = await generateText({
          model: atxp('gpt-4.1'),
          prompt: 'What is the capital of France?',
        });

        // Process the result
        console.log(text);
        ```
      </Tab>

      <Tab title="Python OpenAI SDK">
        Install the OpenAI SDK and requests package in your project.

        ```bash theme={null}
        pip install openai requests
        ```

        Create an OpenAI client and use it.

        ```python theme={null}
        # Import the OpenAI SDK
        from openai import OpenAI

        # Create the client
        client = OpenAI(
            api_key=os.environ.get("ATXP_CONNECTION"),
            base_url=API_BASE_URL,
        )

        # Chat with the LLM
        completion = client.chat.completions.create(
            model="gpt-4.1",
            messages=[
                {"role": "system", "content": "You are a helpful assistant."},
                {"role": "user", "content": "What is 2+2?"}
            ]
        )

        # Process the result
        print(completion.choices[0].message.content)
        ```
      </Tab>

      <Tab title="HTTP">
        Call the API directly via HTTP requests. The structure is the same as the OpenAI [chat completions](https://platform.openai.com/docs/api-reference/chat) API.

        ```bash theme={null}
        curl -X POST "https://llm.atxp.ai/v1/chat/completions" \
          -H "Authorization: Bearer $ATXP_CONNECTION" \
          -H "Content-Type: application/json" \
          -d '{
            "model": "gpt-4.1",
            "messages": [
              {"role": "user", "content": "Say hello!"}
            ],
            "stream": false
          }'
        ```
      </Tab>
    </Tabs>
  </Step>
</Steps>