In recent years, deep learning and simulation have come to receive a significant amount of focus from researchers. A great deal of work has gone into connecting these two fields, however, merging them is not something that can be accomplished quickly. The majority of those possible combinations include using the results of the simulation as input for deep learning, or vice versa. There are not yet many possible mixtures in which the agent possesses an innate capacity for deep learning, which would contribute to the agent’s more intelligent behavior. In addition, the majority of deep learning models are built with Python, whereas the GAMA platform [1] and NetLogo [2] are frequently used to create simulation models. As a means of bridging these two domains, we present Gampy - a GAMA plugin that assists in the integration of a deep-learning model that has been pre-trained in Python. The agent has been updated to provide a predictive capability that makes use of models trained from Python deep learning frameworks (Tensorflow/Keras [3, 4] and Pytorch [5])). Moreover, we demonstrate the efficiency of this plugin by showcasing a few different applications. The findings reveal that Gampy’s execution time outperforms alternative networking approaches.