The rapid advancement of AI-based language models has transformed the field of Natural Language Processing (NLP) into a powerful tool for text generation. This study evaluates the performance of models in different categories such as factual accuracy, creative writing, open-ended writing, and technical explanation. We have considered three popular and advanced large language models (LLMs) for this analysis. To quantify their performance, we have applied a combination of statistical and linguistic metrics. We have used Dale-Chall to analyze the readability score of the responses. For lexical diversity, we have used the type-token ratio technique. In addition, a cosine similarity with TF-IDF i... More >