Recent advancements in deep learning, especially the Transformer architectures, have emerged as the current state-of-the-art natural language processing (NLP). The generalizability and few-shot capabilities of large language models (LLM) base on Transformer architecture like GPT-3 have opened up new possibilities for countless innovative apps across various industries, including healthcare. LLMs demonstrate an impressive context and language understanding that enables them to solve problems that were previously intractable with deep learning. And more importantly, the learning context and target is not limited to human understandable language. In this talk, we will share how healthcare industry leverage LLM to solve various domain context and problems, such as electronic healthcare record (EHR), protein structure and also the tools/SDKs that Nvidia develop and provide to the field helping industry accelerate LLM model development.