a AI powered chatbot that takes your pdfs and text files process it and gives result from it. Used LLama.cpp and gradio with llms and mongodb as a backend to store vector databse. used tesseract ocr and pymypdf for image and pdf processing. It also store the conversion in memeory for a conversational results.
✨ Key Features:
Text & PDF Processing using Tesseract OCR and PyMuPDF for seamless handling of images and documents. Powered by LLama.cpp and Llama2 13B Chat GGUF model from HuggingFace for state-of-the-art natural language processing. Utilizes FAISS to find similar context in documents, ensuring the chatbot delivers precise, context-aware answers. MongoDB as the backend, storing a vector database to keep track of the data, ensuring fast and reliable queries. The system also stores context in memory, making conversations more interactive and engaging. #Application - UI
** i downloaded the model from huggingface - llama2 13b chat gguf model **
** and used FAISS for finding similiar context from the docs. **