This thesis explored the application of Large Language Models (LLMs) to automate SQL query generation, aiming to bridge the gap between natural language and database interaction. By Exploring different LLMs and analyzing their architecture and their ability to understand different large language and manage to produce a corresponding SQL query that fulfils the task. In specific several LLM were explored to achieve our goal of finding the most suitable LLM for the task and among these models were the Deepseek7B and Mistral7B which have shown more advanced results compared to others. Specifically by separating the problem into 2 tasks we have managed to increase the overall performance such that for the first task it will be related to selecting the most suitable schema for the tables for the asked question and the second task will be for using that of the first task along with the question to get the final output.