Artificial intelligence (AI) has advanced so quickly in recent years that it has changed the way of approach of several sectors, mainly information services. With the use of an AI-driven user-friendly platform, this study seeks to make it easier to address the queries of the users to obtain any type of information about the institution. This project aims to create SASBOT (SASTRA ChatBOT), a novel AI chatbot specifically intended to understand and respond to queries from students at SASTRA University. SASBOT offers question-and-answer capabilities regarding SASTRA University by utilizing advanced closed-source and open-source Large Language Models (LLMs), specifically OpenAI GPT-3.5-turbo and mistral AI’s mixtral-8x7B integrated with the vector database which contains data of SASTRA website collected using LangChain document loaders. In SASBOT, LLMs are responsible for understanding the relation between user queries and retrieved information from the vector database based on similarity search to generate human-like responses. Using OpenAI API and inferencing, SASBOT can process natural language inputs, providing comprehensive and contextually relevant answers to user inquiries. The Hallucination scores of both LLMs are calculated using Hughes Hallucination Evaluation Model to compare the LLM’s performances.