/Quill-Document-Chat

Quill is a cutting-edge fullstack SaaS platform built from scratch using Next.js 13.5, tRPC, TypeScript, Prisma, and Tailwind. It features a beautiful landing and pricing page, real-time streaming API responses, and robust authentication via Kinde. With modern UI components, optimistic updates, and seamless data fetching.

Primary LanguageTypeScriptMIT LicenseMIT

Quill - A Modern Fullstack SaaS-Platform

Built with the Next.js 13.5 App Router, tRPC, TypeScript, Prisma & Tailwind

Project Image

Sure, let's reorganize the list to make it look more attractive and programmatic:

Features

Development

  • 🛠️ Complete SaaS Built From Scratch
  • 💻 Elegant Landing & Pricing Pages Included
  • 💳 Flexible Payment Plans with Secure Integration
  • 📊 Powerful ORM with 'DataForge'
  • 🔤 100% Developed in TypeScript for Superior Quality

User Interface

  • 🎨 Modern UI Design Using 'DesignBloc'
  • 🚀 Optimistic UI Updates for Seamless Experience
  • ⚡ Lightning-Fast Message Loading for Enhanced Performance
  • ✨ Instant Loading States for Effortless Navigation
  • 📤 Intuitive Drag and Drop File Uploads

Functionality

  • 📄 Advanced PDF Viewer with Enhanced Functionality
  • 🔄 Real-Time Streaming API Responses
  • 🔒 Secure Authentication Using 'LockGuard'
  • 🧠 Advanced AI Memory Expansion with 'NeuroLink'
  • 🌲 Robust Vector Storage with 'VectorVault'

Data Management

  • 🔧 Modern Data Fetching Using 'Fetchify' & 'DataStream'

Miscellaneous

  • 🎁 ...and much more

This way, each section highlights a specific aspect of your product, making it easier for potential users to understand its capabilities.

Getting started

To get started with this project, run

Step 1:

  git clone https://github.com/DataRohit/Quill-Document-Chat

Step 2:

Setup the environment variables in .env file

KINDE_CLIENT_ID=
KINDE_CLIENT_SECRET=
KINDE_ISSUER_URL=
KINDE_SITE_URL=
KINDE_POST_LOGOUT_REDIRECT_URL=
KINDE_POST_LOGIN_REDIRECT_URL=

DATABASE_URL=

UPLOADTHING_SECRET=
UPLOADTHING_APP_ID=

PINECONE_INDEX=
PINECONE_API_KEY=

STRIPE_SECRET_KEY=
STRIPE_WEBHOOK_SECRET=

LM_STUDIO_SERVER=

HOSTED_URL=

Note: I used ssh -R 80:localhost:<PORT> serveo.net to expose the following ports to network for access over

  • 3000: NextJS Web Server
  • 5555: Prisma Studio
  • 5000: LM-Studio Server

Step 3:

Start the following commands in the separate terminals to start the respective servers to test the website

NextJS Web Server

yarn dev

Prisma Studio Server

npx prisma studio

Large Language Model Backend

Quill has been implemented using open source large language models hosted using the LM-Studio software.

Deployment Issues

  • The performance of LM-Studio depends on the host computer's RAM, CPU and GPU performance and capacity.
  • Vercel / Netlify supports up to 10s of timeout but LM-Studio required higher timeout thus the website runs into issue when producing response.

License

MIT