📄 Chat With PDF (Local AI)

Chat With PDF is a full-stack application that allows users to upload PDF files and interact with them using local AI models.
The project uses Ollama for local LLM inference, Qdrant as the vector database, Valkey for caching/queueing, and Clerk for authentication.

---

✨ Features

• 🔐 Authentication with Clerk
• 📄 Upload and chat with PDF documents
• 🤖 Local AI using Ollama
• 🧠 Vector search powered by Qdrant
• ⚡ Fast caching with Valkey
• 🐳 Docker-based infrastructure
• 🌐 Full-stack Next.js app

---

🧱 Tech Stack

• Frontend / Backend: Next.js (App Router)
• Authentication: Clerk
• LLM Runtime: Ollama (local)
• Vector Database: Qdrant
• Cache / Queue: Valkey
• Package Manager: npm
• Containerization: Docker

---

📦 Prerequisites

Make sure you have the following installed:

• Node.js >=18
• Docker & Docker Compose
• Ollama
• npm

---

🧠 Install Ollama (Local AI)

Install Ollama:

curl -fsSL https://ollama.com/install.sh | sh

Pull a model (example):

ollama pull llama3

Start Ollama (usually auto-starts):

ollama serve

Ollama will be available at:

http://localhost:11434

---

🐳 Run Required Services (Docker)

1️⃣ Qdrant (Vector Database)

docker run -d \
  --name qdrant \
  -p 6333:6333 \
  -p 6334:6334 \
  qdrant/qdrant

2️⃣ Valkey (Cache)

docker run -d \
  --name valkey \
  -p 6379:6379 \
  valkey/valkey

Verify containers:

docker ps

Expected output (example):

qdrant   0.0.0.0:6333-6334->6333-6334/tcp
valkey   0.0.0.0:6379->6379/tcp

---

🔐 Clerk Authentication Setup

1. Create an account at https://clerk.com
2. Create a new application
3. Copy your keys

Create a .env.local file in the project root:

# Clerk
NEXT_PUBLIC_CLERK_PUBLISHABLE_KEY=pk_test_xxxxxxxxx
CLERK_SECRET_KEY=sk_test_xxxxxxxxx

# App
NEXT_PUBLIC_APP_URL=http://localhost:3000

# Qdrant
QDRANT_URL=http://localhost:6333

# Valkey
VALKEY_URL=redis://localhost:6379

# Ollama
OLLAMA_BASE_URL=http://localhost:11434
OLLAMA_MODEL=llama3

---

📥 Installation

Clone the repository:

git clone https://github.com/tarunkumar-sys/chat_with_pdf.git
cd chat_with_pdf

Install dependencies:

npm install

---

🚀 Running the App

Start the full development environment:

npm run dev:full

The application will be available at:

http://localhost:3000

---

🧪 Troubleshooting

Containers stopped?

Restart them:

docker start qdrant valkey

Ollama not responding?

Check status:

ollama list

Restart:

ollama serve

Port conflicts?

Ensure these ports are free:

• 3000 – App
• 6333 – Qdrant
• 6379 – Valkey
• 11434 – Ollama

---

📄 License

MIT License