In this work, we explore multiple neural architectures adapted for the task automatic post-editing of machine translation output. We focus on neural-to-end models that combine both inputs mt (raw MT output) and src (source language input) in a single neural architecture, modeling \\mt, src\\\ pe directly. Apart from that, we investigate the influence of-attention models which seem to be well-suited for monolingual tasks, as as combinations of both ideas. We report results on data sets provided the WMT-2016 shared task on automatic post-editing and can demonstrate dual-attention models that incorporate all available data in the APE in a single model improve on the best shared task system and on all published results after the shared task. Dual-attention models that are with hard attention remain competitive despite applying fewer changes the input.