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

# Vapi integration guide

Vapi is a developer platform for building voice AI agents. It handles the underlying infrastructure so developers can focus on creating high-quality voice experiences. Voice agents built with Vapi can engage in natural conversations with users, make and receive phone calls, integrate seamlessly with existing systems and APIs, and support complex workflows such as appointment scheduling, customer support, and other advanced use cases.

SambaNova's high-speed inference enables low-latency voice interactions, which is critical for natural-sounding conversations.

## Prerequisites

Before starting, ensure you have:

* A [SambaCloud](https://cloud.sambanova.ai/apis) account and API key.
* A [Vapi](https://dashboard.vapi.ai/login) account with dashboard access.
* Python 3.11 or later.
* [ngrok](https://ngrok.com/) installed for exposing your local server.

## Installation and setup

### Clone the repository

```bash theme={null}
git clone https://github.com/sambanova/integrations.git
cd integrations/vapi
```

### Create a virtual environment

```bash theme={null}
python -m venv .venv
source .venv/bin/activate
```

### Install dependencies

```bash theme={null}
pip install flask==3.1.2 sambanova==1.2.0
```

### Install ngrok

On macOS:

```bash theme={null}
brew install ngrok
```

Then add your ngrok auth token. For more information, see the [ngrok documentation](https://dashboard.ngrok.com/get-started/your-authtoken):

```bash theme={null}
ngrok config add-authtoken $YOUR_NGROK_AUTHTOKEN
```

## Set environment variables

Create a `.env` file or export your API key directly:

```bash theme={null}
export SAMBANOVA_API_KEY="your-sambanova-api-key"
```

You can obtain your API key from the [SambaCloud portal](https://cloud.sambanova.ai/apis).

## Run the local LLM server

Start the Flask server:

```bash theme={null}
python app.py
```

The server will run on `http://localhost:5000/`.

## Expose the server using ngrok

In a separate terminal, run:

```bash theme={null}
ngrok http 5000
```

ngrok will generate a public URL similar to:

```
https://abcd-1234.ngrok-free.dev
```

This is the endpoint Vapi will call.

### Test the endpoint

Verify your setup with a cURL request:

```bash theme={null}
curl -X POST https://abcd-1234.ngrok-free.dev/chat/completions \
  -H "Content-Type: application/json" \
  -d '{
    "call": "chat.completions",
    "metadata": {
      "request_id": "example-123"
    },
    "model": "gpt-oss-120b",
    "messages": [
      {
        "role": "system",
        "content": "You are a helpful assistant."
      },
      {
        "role": "user",
        "content": "Hello! Explain what an LLM is in one sentence."
      }
    ],
    "temperature": 0.7,
    "max_tokens": 150,
    "stream": true
  }'
```

<Note>
  Replace `https://abcd-1234.ngrok-free.dev` with your actual ngrok URL.
</Note>

## Configure Vapi with your custom LLM

1. Log in to the [Vapi Dashboard](https://dashboard.vapi.ai/).

2. Create a new assistant using a **Blank Template**.

3. Navigate to **Model** → **Provider** → **Custom LLM**.

4. Enter the model name you'll use (for example, `gpt-oss-120b`).

5. Paste your ngrok URL into the endpoint URL field:

   ```
   https://abcd-1234.ngrok-free.dev/chat/completions
   ```

6. Save the configuration.

## Additional resources

For detailed instructions and the complete source code, see the [Vapi integration example](https://github.com/sambanova/integrations/tree/main/vapi) on GitHub.

## Vapi documentation

For more information about Vapi, see the [official Vapi documentation](https://docs.vapi.ai/quickstart/introduction).