This work tries to systematically compare two cutting-edge language models, Bio BERT and LLM (large language model) within the framework of developing a novel medical chatbot. The objective of our work is to develop a healthcare information system that will be based on fine-tuned language models and will be able to communicate with users in an accurate, sensitive and customized manner. The model performance of LLaMA, the large model language, and Bio BERT, designed specifically for text data in the biomedical domain are compared in several tasks including entity recognition, question answering, and dialogue generation in the medical area. The results suggest that both models have the ability to comprehend sophisticated biomedical ideas and produce useful replies. The results show that Bio BERT is able to achieve very high effectiveness on named entity recognition and question answering by getting access to a huge number of biomedical concepts and contextual information. On the other hand, Llama is able to generate emotionally and contextually relevant responses increasing user appreciation and involvement. The target goal in performing this comparative analysis is to highlight the merits and demerits of the features of each model as well as assist in guiding future medical research directions. This work utilizes state-of-the-art language models to provide personalized medical services and hence contribute to solutions to issues of digital healthcare and global health literacy. The comparative study also examines the scalability and cost efficiency as well as the practical implementation of both the Bio BERT and Llama models in the healthcare settings. Bio BERT achieves good performance in specific biomedical tasks but due to its high computational needs and fine-tuning procedure it may have limited penetration in the wider market. At the same time, the Llama model has the advantages of being able to be expanded and integrated quite easily which might possibly facilitate the development and the introduction of medical chatbots in a number of healthcare settings. The research aims at answering such questions and in doing so provide a baseline for understanding the potential helpfulness of elegant language models in enhancing healthcare information systems.