Skip to content
TheChatArchiveThe Chat Archive
Geminigemini-2.5-flash-preview-04-17· GoogleHugging Face LMArena Human Preference 140K

create a simple llm chat ui

From a Chatbot Arena battleThe same person sent this prompt to two anonymous models at once. Compare it with o3-mini’s answer.
  1. User

    create a simple llm chat ui

  2. Gemini Assistant

    Okay, here is a very simple LLM chat UI using just HTML, CSS, and JavaScript.

    This example simulates an LLM response by just echoing the user's message. To connect it to a real LLM, you would replace the simulation logic with an API call.

    1. index.html (The Structure)

    <!DOCTYPE html>
    <html lang="en">
    <head>
        <meta charset="UTF-8">
        <meta name="viewport" content="width=device-width, initial-scale=1.0">
        <title>Simple LLM Chat UI</title>
        <link rel="stylesheet" href="style.css">
    </head>
    <body>
        <div id="chat-container">
            <div id="chat-display">
                <!-- Messages will appear here -->
                <div class="message llm-message">Welcome to the simple chat! Type your message below.</div>
            </div>
            <div id="input-area">
                <input type="text" id="user-input" placeholder="Type your message...">
                <button id="send-button">Send</button>
            </div>
        </div>
    
        <script src="script.js"></script>
    </body>
    </html>
    

    2. style.css (The Styling)

    body {
        font-family: sans-serif;
        margin: 0;
        padding: 0;
        background-color: #f0f0f0;
        display: flex;
        justify-content: center;
        align-items: center;
        min-height: 100vh;
    }
    
    #chat-container {
        width: 90%;
        max-width: 600px; /* Max width for larger screens */
        height: 80vh;    /* Use 80% of viewport height */
        max-height: 700px; /* Max height */
        background-color: #fff;
        border-radius: 8px;
        box-shadow: 0 0 15px rgba(0, 0, 0, 0.1);
        display: flex;
        flex-direction: column; /* Stack display and input */
        overflow: hidden; /* Keep contents inside rounded corners */
    }
    
    #chat-display {
        flex-grow: 1; /* Take up all available space */
        padding: 15px;
        overflow-y: auto; /* Make it scrollable */
        display: flex; /* Use flexbox for messages */
        flex-direction: column; /* Stack messages */
    }
    
    .message {
        margin-bottom: 10px;
        padding: 10px 15px;
        border-radius: 18px; /* More rounded corners */
        max-width: 80%; /* Don't let messages take full width */
        word-wrap: break-word; /* Prevent long words overflowing */
    }
    
    .user-message {
        background-color: #007bff; /* Blue background for user */
        color: white;
        align-self: flex-end; /* Align to the right */
        margin-left: auto; /* Push to the right */
    }
    
    .llm-message {
        background-color: #e9e9eb; /* Light grey background for LLM */
        color: #333;
        align-self: flex-start; /* Align to the left */
        margin-right: auto; /* Push to the left */
    }
    
    #input-area {
        display: flex; /* Arrange input and button side-by-side */
        padding: 15px;
        background-color: #f9f9f9;
        border-top: 1px solid #eee;
    }
    
    #user-input {
        flex-grow: 1; /* Take up available space */
        padding: 10px;
        border: 1px solid #ccc;
        border-radius: 20px; /* Rounded input field */
        margin-right: 10px;
        font-size: 1rem;
        outline: none; /* Remove default outline */
    }
    
    #user-input:focus {
        border-color: #007bff; /* Highlight on focus */
        box-shadow: 0 0 5px rgba(0, 123, 255, 0.25);
    }
    
    #send-button {
        padding: 10px 20px;
        background-color: #28a745; /* Green send button */
        color: white;
        border: none;
        border-radius: 20px; /* Rounded button */
        cursor: pointer;
        font-size: 1rem;
        transition: background-color 0.2s ease;
    }
    
    #send-button:hover {
        background-color: #218838; /* Darker green on hover */
    }
    
    #send-button:active {
         background-color: #1e7e34; /* Even darker when clicked */
    }
    

    3. script.js (The Logic)

    document.addEventListener('DOMContentLoaded', () => {
        const userInput = document.getElementById('user-input');
        const sendButton = document.getElementById('send-button');
        const chatDisplay = document.getElementById('chat-display');
    
        // Function to add a message to the chat display
        function addMessage(text, sender) {
            const messageDiv = document.createElement('div');
            messageDiv.classList.add('message');
            messageDiv.classList.add(`${sender}-message`); // Adds 'user-message' or 'llm-message'
            messageDiv.textContent = text;
            chatDisplay.appendChild(messageDiv);
    
            // Auto-scroll to the bottom
            chatDisplay.scrollTop = chatDisplay.scrollHeight;
        }
    
        // Function to simulate LLM response (replace with actual API call)
        function getLLMResponse(userMessage) {
            // --- This is where you would integrate with an LLM API ---
            // For this simple example, we just echo the message
            const simulatedResponse = `You said: "${userMessage}". (This is a simulated LLM response)`;
    
            // Simulate a delay for realism
            setTimeout(() => {
                addMessage(simulatedResponse, 'llm');
            }, 500); // Wait 500 milliseconds
            // -----------------------------------------------------------
        }
    
        // Function to handle sending a message
        function sendMessage() {
            const messageText = userInput.value.trim(); // Get text and remove whitespace
    
            if (messageText === '') {
                return; // Don't send empty messages
            }
    
            // 1. Display user message
            addMessage(messageText, 'user');
    
            // 2. Get and display LLM response (simulated here)
            getLLMResponse(messageText);
    
            // 3. Clear the input field
            userInput.value = '';
        }
    
        // Event listener for the Send button click
        sendButton.addEventListener('click', sendMessage);
    
        // Event listener for pressing Enter key in the input field
        userInput.addEventListener('keypress', (event) => {
            // Check if the key pressed was 'Enter' (key code 13)
            if (event.key === 'Enter') {
                event.preventDefault(); // Prevent default form submission (if input was in a form)
                sendMessage();
            }
        });
    });
    

    How to Use:

    1. Save the three code blocks above into three separate files: index.html, style.css, and script.js.
    2. Make sure all three files are in the same folder.
    3. Open the index.html file in your web browser.

    You will see a simple chat interface. Type into the box, click "Send" or press Enter, and your message will appear on the right (blue), followed by a simulated response from the "LLM" on the left (grey).

    To integrate with a real LLM:

    1. Sign up for an LLM API (e.g., OpenAI, Google AI, Anthropic, etc.).
    2. Get your API key.
    3. In the getLLMResponse function in script.js, replace the simulation logic (const simulatedResponse = ...; setTimeout(...)) with code that makes an HTTP request to the LLM provider's API, sending the userMessage.
    4. Handle the API response (which will be the LLM's text) and pass it to addMessage(llmResponseText, 'llm');.
    5. Important: For a client-side UI like this, you should generally not put your API key directly in the client-side JavaScript. A more secure approach is to have a small backend server (using Node.js, Python/Flask, etc.) that receives the user message from the frontend, makes the API call securely using your key, and then sends the LLM response back to the frontend.

Responses were produced in different, uncontrolled contexts (system prompts, settings, prior turns). This is not a benchmark.

Identical first prompt