Finding sarcasm involves determining whether a sentence is intended to be ironic or sarcastic. Natural language processing is a difficult task since sarcasm usually uses language that isn’t meant to be taken literally. To address this issue, experimenters have recently created a number of strategies, including rule-based techniques, machine learning models, and deep learning architectures. These approaches generally rely on features such as sentiment, context, and linguistic cues to identify sarcastic statements. While significant progress has been made in sarcasm detection, it remains an open research question due to the complex nature of sarcasm and the subjective nature of its interpretation. The model’s performance is compared to several traditional and modern methods, including attention-based models, Support Vector Machines, Logistic Regression, Long Short Term Memory, and Recurrent Neural Networks, which have been claimed to be employed for related tasks. The proposed model establishes its competency through evaluation on many parameters such as accuracy, precision, recall, and F1 score.