Breast cancer remains one of the most prevalent and deadly forms of cancer worldwide, affecting individuals across all ages and sexes. In recent years, artificial intelligence (AI) and machine learning (ML) have emerged as transformative tools in the field of medical diagnostics, offering the potential to enhance the detection, diagnosis, and prediction of breast cancer. Despite these advancements, challenges remain, including the need for large, diverse datasets to train robust models, the integration of AI tools into clinical workflows, and addressing ethical concerns related to AI in healthcare. This paper explores the application of Large Language Models (LLMs) using embeddings in breast cancer management, focusing on its ability to analyze medical data, including imaging, histopathology datasets to identify patterns that may be imperceptible to human experts. Datasets from real-world setting have been secured for analysis across multiple models. Convolutional Neural Network (CNN) model and custom-built large language model are employed to demonstrate the precision and accuracy of Generative AI techniques and observed that custom-built LLM with 98.44% outperforms the traditional AI approaches such as CNN with 61.72%. Future studies can further establish how these models can assist in stratifying patients based on risk, thereby enabling personalized treatment plans that can reduce overtreatment and improve quality of life.