In this project, we have developed an efficient, offline-ready, advanced LLM (Large Language Model) with low GPU/CPU usage, high accuracy, versatile conversation handling, document uploads, secure data handling, user-selectable models, and customizable response lengths. Developers can obtain the mojo file separately and integrate it into their applications, significantly boosting their chatbot's speed, up to 2000 times faster.
This project is built using MOJO, a new programming language for AI/ML and for web page view it is using python DJango for back-end.
As all we know that python is slow, we need an alternative for that which should have all resources as python and faster than this. So that's how MOJO came into the picture, it has all libraries which python has and much faster than python (35000x).
You can integrate this LLM with any of your applications which uses a language model.
PROCEDURES TO FOLLOW:
---> Follow the steps & Install MOJO as per official documentation. (https://developer.modular.com/download)
---> Install python from python.org.
---> Now Fork the Github Repo.
---> Download the bin file from this link (https://drive.google.com/file/d/1NftSxE7iu0sndEDq51BVMDvYW7Hgy2px/view?usp=sharing) or download any other model.
---> As DJango has been used, run this in the terminal (pip install django)
---> Run the LLM with the command (python3 manage.py runserver).