This paper focuses on NMT systems, NMT is developed with deep learning techniques; it's kind of MT attempts to build and train a single, large neural network that reads a sentence and outputs a correct translation. So, we proposed a system on based NMT systems for the translation of Arabic texts into English, This system is for translating texts in the computer field, The proposed system contains five main stages , the system is contains in a database containing a bilingual dictionary (Arabic-English) containing terms in the field of Computer-based, this system based on the encoder-decoder by RNN model in which the encoder reads and encodes the source sentence into a fixed-length vector while the decoder produces a translation output from the encoded vector, the system was evaluated by human experts , addition to using the blue scale, The final results proved that the proposed method achieved higher performance than other systems.