Determination of spinach plant predictions is one of the most critical decision-making processes. In predicting spinach plants in each period, it depends on each period, both the previous and subsequent periods. The production of spinach plants that change every period causes uncertainty in predicting. The method used to indicate the data is the Fletcher-Reeves algorithm, it is an appropriate development technique compared to the backpropagation strategy because this strategy can speed up the preparation time to arrive at the minimum convergence value. This paper does not discuss the prediction results. Still, it discusses the ability of the Fletcher-Reeves algorithm to make predictions based on the spinach production dataset obtained from the Central Statistics Agency. The purpose of this research is to see the accuracy and performance measurement of the algorithm in the search for the best results to solve the prediction of spinach plants in Sumatra. The research data used are spinach vegetable production data in North Sumatra. Based on this data, a network architecture model will be formed and determined, including 2-20-1, 2-30-1, 2-35-1, 2-45-1, and 2-50-1. After training and testing, these five models show that the best architectural model is 2-20-1 with an MSE value of 0.00608399, the lowest among the other four models. So the model can be used to predict spinach plants in Sumatra.A well-prepared abstract enables the reader to identify the basic content of a document quickly and accurately, to determine its relevance to their interests, and thus to decide whether to read the document in its entirety.