Leveraging the potential of large language models