The rapid growth of textual data in the digital age has necessitated the development of efficient Text Summarization systems to distill critical information from extensive documents. This study presents a metric-driven comprehensive comparative analysis of three state-of-the-art Large Language Models (LLMs) for Text Summarization: ChatGPT-4o, Gemini 1.5 Pro, and Cohere Command. By adopting sophisticated evaluation measures like ROUGE, BERTScore, METEOR, and Cosine Similarity, the effectiveness of these algorithms in producing precise and cohesive summaries is evaluated. The findings show that ChatGPT-4o continuously performs better than its counterparts in terms of semantic similarity and structural coherence, even if all three models have noteworthy capabilities. The study highlights how LLMs can improve text summarization tasks and offers important insights into their strengths and limitations.