Lexara AI - Modern AI Chatbot

A beautiful, modern web-based AI chatbot that replicates the core user experience and interface of ChatGPT, powered by the Google Gemini API.
Visit Now

Features

• Modern Aesthetic Design: Clean, beautiful interface inspired by the latest AI chat applications.
• Lexara AI Branding: Unique identity with custom logo and gradient design.
• Real-time Messaging: Live chat with elegant typing indicators.
• Quick Action Cards: Fast access to common AI tasks (with modern Fluent design).
• Code Syntax Highlighting: Beautiful formatting for code snippets using highlight.js.
• Markdown & Clean Typography: Completely styled headers, lists, paragraphs, horizontal rules, links, blockquotes, and tables for a highly professional look.
• Responsive & Mobile-First Design: Fully optimized for desktop, tablet, and mobile (with mobile drawer navigation, touch-friendly controls, and a mobile-optimized theme toggle).
• Persistent Chat History: All your chats are saved in your browser's local storage, so you can revisit previous conversations even after a reload.
• Chat Deletion: Delete individual chats easily via the trashcan icon next to any thread in the history sidebar.
• Clipboard Sharing: One-click button to copy your entire conversation to the clipboard.
• Refined Theme Toggle: Modern, touch-friendly dark/light mode toggle with smooth transitions and custom SVG icons.
• Conversation Context & System Instruction: Maintains context throughout the session. Includes system prompt guiding for knowledge limits gracefully.

Technology Stack

• Frontend: HTML, CSS, vanilla JavaScript
• Backend: Python with FastAPI
• AI Model: Google Gemini API (gemini-2.5-flash by default)
• Dependencies: OpenAI library (configured for Gemini's OpenAI-compatible endpoint)

Setup Instructions

Prerequisites

• Python 3.8 or higher
• Gemini API key (get one from Google AI Studio)

Installation

1. Clone or download the project files

2. Install Python dependencies:

   pip install -r requirements.txt

3. Set up environment variables:

   • Open the .env file
   • Replace with your actual Gemini API key and model settings:

     GEMINI_API_KEY=your_gemini_api_key_here
     GEMINI_MODEL=gemini-2.5-flash

4. Run the application:

   python main.py

5. Open your browser and navigate to:

   http://localhost:8000
   

6. Deployed using Heroku

Project Structure

lexara-ai/
├── main.py                 # FastAPI backend server
├── requirements.txt        # Python dependencies
├── .env                   # Environment variables (API key)
├── README.md              # This file
├── run.bat                # Windows batch file for quick start
├── setup_and_run.py       # Python setup helper
└── static/                # Frontend files
   ├── index.html         # Main HTML page
   ├── styles.css         # CSS styling (including mobile, dark mode, and markdown)
   └── script.js          # JavaScript (UI logic, chat history, and deletion)

API Endpoints

• GET / - Serves the main HTML page
• POST /api/chat - Handles chat requests
• GET /health - Health check endpoint

Customization

Styling

Modify static/styles.css to change the appearance:

• Colors and themes
• Layout and spacing
• Markdown/typography overrides

Functionality

Extend static/script.js to add new features:

• Local storage chat history (see saveCurrentChat, loadChatHistoryFromStorage, deleteChat)
• Clipboard sharing (see copyChatToClipboard)
• Custom commands

Getting Your Gemini API Key

1. Visit Google AI Studio
2. Sign up or log in with your Google account
3. Click Get API key
4. Copy the API key to your .env file as GEMINI_API_KEY

Troubleshooting

Common Issues

1. "Module not found" errors:

   • Make sure you've installed all requirements: pip install -r requirements.txt

2. API key errors / 403 / 429:

   • Verify your Gemini API key is correctly set in the .env file.
   • Note that Google API keys generated outside AI Studio or restricted projects may return 403 Permission Denied or 429 Quota Exceeded errors. Generate a standard key starting with AIzaSy.

3. Port already in use:

   • Change the port in main.py: uvicorn.run(app, host="0.0.0.0", port=8001)

License

This project is for educational purposes.