Chat with PDF

A simple RAG (Retrieval-Augmented Generation) system using Deepseek, LangChain, and Streamlit to chat with PDFs and answer complex questions about your local documents.

You can watch the video on how it was built on my YouTube.

Pre-requisites

Install Ollama on your local machine from the official website. And then pull the Deepseek model:

ollama pull deepseek-r1:14b

Install the dependencies using pip:

pip install -r requirements.txt

Run

Run the Streamlit app:

streamlit run pdf_rag.py