Building a Chat Application with Ollama's Llama 3 Model Using JavaScript, HTML, and CSS

Introduction

In this blog post, we'll walk through the process of creating a simple chat application that interacts with Ollama's Llama 3 model. We'll use JavaScript, HTML, and CSS for the frontend, and Node.js with Express for the backend. By the end, you'll have a working chat application that sends user messages to the AI model and displays the responses in real-time.

Prerequisites

Before you begin, ensure you have the following installed on your machine:

Node.js
npm (Node Package Manager)

Step 1: Setting Up the Frontend

HTML

First, create an HTML file named index.html that defines the structure of our chat application.

<!DOCTYPE html>
<html lang="en">
<head>
    <meta charset="UTF-8">
    <meta name="viewport" content="width=device-width, initial-scale=1.0">
    <title>Chat with Ollama's Llama 3</title>
    <link rel="stylesheet" href="styles.css">
</head>
<body>
    <div id="chat-container">
        <div id="chat-window">
            <div id="messages"></div>
        </div>
        <input type="text" id="user-input" placeholder="Type your message here...">
        <button id="send-button">Send</button>
    </div>
    <script src="script.js"></script>
</body>
</html>

This HTML file includes a container for the chat messages, an input field for user messages, and a send button.

CSS

Next, create a CSS file named styles.css to style the chat application.

body {
    font-family: Arial, sans-serif;
    display: flex;
    justify-content: center;
    align-items: center;
    height: 100vh;
    background-color: #f0f0f0;
    margin: 0;
}

#chat-container {
    width: 400px;
    border: 1px solid #ccc;
    background-color: #fff;
    border-radius: 8px;
    box-shadow: 0 4px 8px rgba(0, 0, 0, 0.1);
    overflow: hidden;
}

#chat-window {
    height: 300px;
    padding: 10px;
    overflow-y: auto;
    border-bottom: 1px solid #ccc;
}

#messages {
    display: flex;
    flex-direction: column;
}

.message {
    padding: 8px;
    margin: 4px 0;
    border-radius: 4px;
}

.user-message {
    align-self: flex-end;
    background-color: #007bff;
    color: #fff;
}

.ai-message {
    align-self: flex-start;
    background-color: #e0e0e0;
    color: #000;
}

#user-input {
    width: calc(100% - 60px);
    padding: 10px;
    border: none;
    border-radius: 0;
    outline: none;
}

#send-button {
    width: 60px;
    padding: 10px;
    border: none;
    background-color: #007bff;
    color: #fff;
    cursor: pointer;
}

This CSS file ensures the chat application looks clean and modern.

JavaScript

Create a JavaScript file named script.js to handle the frontend functionality.

document.getElementById('send-button').addEventListener('click', sendMessage);
document.getElementById('user-input').addEventListener('keypress', function (e) {
    if (e.key === 'Enter') {
        sendMessage();
    }
});

function sendMessage() {
    const userInput = document.getElementById('user-input');
    const messageText = userInput.value.trim();

    if (messageText === '') return;

    displayMessage(messageText, 'user-message');
    userInput.value = '';

    // Send the message to the local AI and get the response
    getAIResponse(messageText).then(aiResponse => {
        displayMessage(aiResponse, 'ai-message');
    }).catch(error => {
        console.error('Error:', error);
        displayMessage('Sorry, something went wrong.', 'ai-message');
    });
}

function displayMessage(text, className) {
    const messageElement = document.createElement('div');
    messageElement.textContent = text;
    messageElement.className = `message ${className}`;
    document.getElementById('messages').appendChild(messageElement);
    document.getElementById('messages').scrollTop = document.getElementById('messages').scrollHeight;
}

async function getAIResponse(userMessage) {
    // Example AJAX call to a local server interacting with Ollama Llama 3
    const response = await fetch('http://localhost:5000/ollama', {
        method: 'POST',
        headers: {
            'Content-Type': 'application/json',
        },
        body: JSON.stringify({ message: userMessage }),
    });

    if (!response.ok) {
        throw new Error('Network response was not ok');
    }

    const data = await response.json();
    return data.response; // Adjust this based on your server's response structure
}

This JavaScript file adds event listeners to the send button and input field, sends user messages to the backend, and displays both user and AI responses.

Step 2: Setting Up the Backend

Node.js and Express

Ensure you have Node.js installed. Then, create a server.js file for the backend.

Install Express:
```
npm install express body-parser
```

Create the server.js file:

const express = require('express');
const bodyParser = require('body-parser');
const app = express();
const port = 5000;

app.use(bodyParser.json());

app.post('/ollama', async (req, res) => {
    const userMessage = req.body.message;

    // Replace this with actual interaction with Ollama's Llama 3
    // This is a placeholder for demonstration purposes
    const aiResponse = await getLlama3Response(userMessage);

    res.json({ response: aiResponse });
});

// Placeholder function to simulate AI response
async function getLlama3Response(userMessage) {
    // Replace this with actual API call to Ollama's Llama 3
    return `Llama 3 says: ${userMessage}`;
}

app.listen(port, () => {
    console.log(`Server running at http://localhost:${port}`);
});

Run the server:
```
node server.js
```

In this setup, your Node.js server will handle incoming requests, interact with Ollama's Llama 3 model, and return responses.

Conclusion

By following these steps, you've created a chat application that sends user messages to Ollama's Llama 3 model and displays the responses. This setup can be extended and customized based on your specific requirements and the features offered by the Llama 3 model.

Feel free to explore and enhance the functionality of your chat application. Happy coding!

Blog