Document summarising has developed as a strong approach for assisting chemistry researchers in obtaining crucial information from a growing corpus of scientific publications. This abstract investigates document summarization's uses in chemical research, emphasising its importance in expediting information acquisition, enhancing decision-making processes, and stimulating creativity. Document summary, in the context of a literature review, allows chemists to sift through a broad array of research papers, extracting essential results and trends, facilitating a thorough grasp of the most recent achievements in their respective disciplines. It speeds up the discovery of pertinent information and knowledge gaps, assisting researchers in developing fresh ideas and performing targeted studies. Document summary greatly aids the pharmaceutical sector by facilitating drug research and development procedures. Researchers can quickly extract critical information from scientific publications about medication targets, chemicals, and biological processes. As a result, promising drug candidates and their mechanisms of action may be identified quickly, expediting drug design and development activities. In this paper, the author investigated a number of pre-trained models, including BERT, GPT-2, BART, T5, and XLNet. Based on an evaluation of their ROUGE scores, the XLNet model surpasses other models in terms of ROUGE score accuracy of more than 60%. As a result, the XLNet model is more suited for document summarization since it creates an accurate summary that includes relevant details about the document or article. To summarise, document summary has transformed chemical research by increasing productivity and encouraging the investigation of innovative ideas. Its capacity to extract crucial insights from massive volumes of textual data enables researchers to progress their fields of study, making it a must-have tool for the modern chemist.