Future frame prediction in videos can be done in two ways: autoregressive and non-autoregressive. Autoregressive methods are accurate at first but decline over time, while non-autoregressive methods lack correlation between predictions. The Implicit Stacked Autoregressive Model for Video Prediction (IAM4VP) combines the best of both methods by using a stacked autoregressive approach. This model outperformed others in predicting future frames in videos, as shown in experiments on various datasets.