In this new era, where tremendous information is offered on the net, it's most significant to supply an improved mechanism to extract the knowledge quickly and most efficiently. it's very difficult for kinsmen to manually extract the summary of an oversized document of text. It is very difficult for human beings to manually extract the summary of a large document of text. Text summarization solves the problem of condensing the information into a more compact form while maintaining the important information in the text. In this thesis, we will discuss a paradigm to perform a review of the abstractive and extractive techniques of text summarization and also will present a new approach/better algorithms towards the problem. Specifically, we perform text summarization on documents like stories, amazon reviews, and Wikipedia dataset using abstractive summarization techniques like RNN based attention mechanisms to produce text summaries of the said documents. The main contribution of the thesis is to provide an insight into the best method for using text summarization.