Chat with Multiple PDFs Using Llama 2 and LangChain - Use Private LLM and Free Embeddings for QA

Learn how to build a chatbot capable of answering questions from multiple PDFs using a private LLM in this comprehensive video tutorial. Utilize the latest Llama 2 13B GPTQ model and LangChain library to create a chain that retrieves relevant documents and generates answers. Discover techniques for loading a GPTQ model with AutoGPTQ, converting PDF directories into vector stores, and implementing text chunk processing. Follow along as the instructor demonstrates the entire process, from setting up Google Colab to preparing the vector database with Instructor Embeddings and creating a functional chain with Llama 2 13B GPTQ. Gain hands-on experience chatting with PDF files and explore practical applications of this technology using real-world examples such as Tesla, Meta, and Nvidia earnings reports.

- Introduction
- Text Tutorial on MLExpert
- Earning Reports PDF Files
- Llama 2 GPTQ
- Google Colab Setup
- Prepare the Vector Database with Instructor Embeddings
- Create a Chain with Llama 2 13B GPTQ
- Chat with PDF Files
- Conclusion