The Indian judiciary system, with its vast backlog of pending cases, faces significant challenges in efficiently managing and delivering timely justice. Case summarization offers a promising solution to expedite legal processes by automatically generating concise summaries of complex legal documents. This research focuses on leveraging Large Language Models (LLMs), specifically T5 and Hugging Face models, to develop an automated case summarization system tailored for the Indian judiciary. Our objective is to create an AI-based system that can accurately summarize judicial case files while preserving the essential legal context and nuances. The proposed approach utilizes pre-trained transformer models, fine-tuned on a legal dataset, to generate coherent and contextually relevant summaries. The models are evaluated based on their ability to reduce document length while maintaining factual accuracy and relevance. Key findings indicate that the transformer-based models demonstrate high efficiency in processing lengthy legal texts, significantly reducing manual workload and improving accessibility to case information. The summarized outputs can be used by legal professionals to quickly grasp the essence of cases, aiding in faster decision-making and reducing case backlog. This research contributes to the ongoing efforts to modernize the legal framework in India, offering a scalable solution that can be integrated into digital judicial platforms, thus enhancing the overall efficiency of the judiciary.