🤖 NVIDIA NIM Demo Chatbot

A Streamlit-based chatbot that uses NVIDIA NIM APIs for document embeddings and large language model (LLM) responses. It allows users to upload PDFs, generate embeddings with NVIDIA's endpoint, and ask contextual questions powered by Llama 3.3-70B Instruct.

---

🚀 Features

• 🔍 Upload and process PDF files from the ./data directory
• 🧠 Create document embeddings using NVIDIAEmbeddings
• 💬 Ask questions based on document context
• 🧾 View retrieved context documents
• 📊 Token and chunk statistics for debugging and transparency
• ⚡ Powered by meta/llama-3.3-70b-instruct via NVIDIA's NIM APIs

---

🛠️ Tech Stack

• Streamlit
• LangChain
• NVIDIA NIM
• FAISS for vector storage
• tiktoken for token counting
• .env for environment variable management

---

📁 Directory Structure

.
├── data/ # PDF files go here
├── .env # NVIDIA_API_KEY goes here
├── app.py # Main Streamlit app
├── requirements.txt # Python dependencies
└── README.md # You're here!

---

🧪 How It Works

1. Load PDFs from the ./data folder
2. Split text into chunks using RecursiveCharacterTextSplitter
3. Filter out chunks exceeding token limits
4. Generate embeddings using NVIDIAEmbeddings
5. Store embeddings in FAISS vector store
6. Ask questions, and the app retrieves relevant chunks and responds using LLM

---

🧑‍💻 Run the App Locally

Follow these steps to run the app in your local environment:

1. 📦 Create and activate a virtual environment

On macOS/Linux:

python3 -m venv venv
source venv/bin/activate

On Windows:

python -m venv venv
venv\Scripts\activate

2. 📥 Install dependencies

Make sure you're in the root directory of the project and run:

pip install -r requirements.txt

3. 🔐 Set your NVIDIA API key

Create a .env file in the root directory with the following content:

NVIDIA_API_KEY=your_nvidia_api_key_here

Make sure to replace your_nvidia_api_key_here with your actual API key from NVIDIA NIM.

4. 🧾 Add your PDF documents

Place your PDF files inside the data/folder. The app will automatically load and process them.

5. 🚀 Run the Streamlit app

Now launch the app using:

streamlit run app.py

The app will open in your default browser. You can:

• Click the "Document Embedding" button to process and embed documents.
• Enter a question in the input box to query your documents using LLM.
• View the retrieved context documents and response time.