Free Course Slides RAG

A fully free local RAG system for course slides

It lets you:

• upload PDF lecture slides
• extract and chunk their text
• embed them with a free sentence-transformers model
• search them with FAISS
• answer questions using a free local Ollama model

Stack

• Streamlit
• PyMuPDF
• sentence-transformers
• FAISS
• Ollama

Requirements

• Python 3.10+
• Ollama installed locally

1) Install Ollama

Install Ollama from the official site, then pull a model:

ollama pull llama3.1:8b

You can also use other local models if your machine supports them.

2) Install Python packages

pip install -r requirements.txt

3) Run the app

streamlit run app.py

How it works

1. Upload one or more PDF slide files.
2. Click Build / Rebuild Index.
3. Ask a question about the slides.
4. The app retrieves the most relevant chunks.
5. If Ollama is running, it generates an answer grounded in the retrieved slide content.
6. If Ollama is not running, the app still shows the retrieved evidence.

Notes

• Best with text-based PDFs.
• If your slides are scanned images, you will need OCR for stronger results.
• The vector index and uploaded files are stored locally in rag_data/.

Suggested improvements

You can extend this project with:

• OCR for image-only slides
• citation formatting in the final answer
• chat history
• Arabic/English answer mode
• per-course folders
• exportable study notes
• support for DOCX or PPTX

Project structure

free_course_slides_rag/
├── app.py
├── requirements.txt
└── README.md