ChatAI

A ChatAI is a perplexity clone with multi-model support, integrating content scraping, source citation to mitigate hallucinations, and rate-limiting for scalability. Designed for accuracy, reliability, and a seamless AI-powered conversational experience.

Built With

• Node.js, Next.js
• Puppeteer, Cheerio
• Upstash Redis
• Groq API
• OpenAI API

Live Demo

Live Site

Getting Started

Prerequisites

• Node.js (v14 or later)
• API keys for Groq, Upstash Redis, and OpenAI

Setup

1. Clone the repository:

   git clone https://github.com/raoakif/chatai.git

2. Navigate to the project directory:

   cd chatai

3. Install dependencies:

   npm install

4. Configure environment variables by creating a .env file with the following content:

   GROQ_API_KEY=""
   UPSTASH_REDIS_REST_URL=""
   UPSTASH_REDIS_REST_TOKEN=""

5. Start the development server:

   npm run dev

6. Access the app at http://localhost:3000.

---

Deployment

To deploy ChatAI to Vercel, follow these steps:

1. Fork the Repository
   Fork the ChatAI repository to your GitHub account.

2. Create a New Project on Vercel

   • Go to Vercel and create a new project from your forked repository.
   • Vercel will automatically detect the repository settings and install the necessary dependencies.

3. Configure Environment Variables
   In the Vercel dashboard, go to the project settings and add the following environment variables:

   GROQ_API_KEY=""
   UPSTASH_REDIS_REST_URL=""
   UPSTASH_REDIS_REST_TOKEN=""

4. Deploy the Project
   Click the Deploy button, and Vercel will automatically build and deploy your project. Once deployed, your ChatAI instance will be accessible via a Vercel-generated URL.

For more detailed deployment instructions, refer to the Vercel Documentation.

---

Features

• Query Handling: Answer queries with context-rich, Markdown-formatted responses.
• Content Scraping: Utilize Puppeteer and Cheerio for web scraping to extract relevant content.
• Rate Limiting: Implement Upstash Redis for smooth handling of user requests.
• Source Citation: Reduce hallucinations by citing sources in answers.

---

Core Files and TODOs

Key Files

• src/app/page.tsx
  Handle dynamic API responses and display user and AI chat interactions.

• src/app/api/chat/route.ts
  Implement the chat API with Groq integration and web scraping using Puppeteer.

• src/middleware.ts
  Manage rate-limiting logic via Upstash Redis.

Planned Enhancements

• Content Scraping Expansion:
  Add support for scraping YouTube videos, PDFs, CSV files, and images.

• Data Visualization:
  Integrate charts (e.g., bar charts, line charts) for enriched responses.

• Hierarchical Web Crawling:
  Develop a crawler to scrape linked content from relevant pages.

---

Learn More

• Next.js Documentation
• Learn Next.js

---

Authors

👤 Rao Akif

• GitHub: @raoakif
• Twitter: @raoakif
• LinkedIn: RaoAkif

🤝 Contributing

Contributions, issues, and feature requests are welcome! Check the issues page for ongoing issues.

Show Your Support

Give a ⭐️ if you like this project!

Acknowledgments

• Inspiration from Perplexity.ai
• Thanks to all contributors and the libraries used in this project.

📝 License

This project is licensed under the MIT License.