This study aims to determine the optimal N, P, K, Mg and Zn rates for groundnut production on Ferric and Plintic Luvisol in the Sudano-Guinean and Sudanian zones of Benin Republic. Two years (2018 and 2019) experiment was carried out in the municipality of Ouessè in the Sudano-Guinean zone and Bembèrèkè in the Sudanian zone. The tested nutrient doses were N (0, 20 and 40 kg⋅ha−1), P (0, 25 and 50 kg⋅ha−1), K (0, 20 and 40 kg⋅ha−1), Mg (0, 15 and 30 kg⋅ha−1) and Zn (0, 4 and 8 kg⋅ha−1). The Box and Behnken rotating design is used to define the N, P, K, Mg and Zn rate combinations leading to 46 combinations. A completely randomized bloc design was setting up considering farmers as replication. In total, four farmers’ fields were selected. A one-way analysis of variance is carried out on yield data, using the linear mixed-effect model. Response surface analyses were used to determine the optimal doses for each N, P, K, Mg and Zn. Nodule production (6.5 times higher than the control), number of gynophores (2.8 times higher than the control) and root length (19.2 ± 0.2 cm) of groundnut plants were significantly (p = 0.0001) improved with nutrient application. The response surface analysis shows that treatments N-P-K-Mg-Zn of 16.01-20.18-6.70-5.65-2.47 (in the Sudano-Guinean zone) and 13.1-25.07-11.47-0-1.82 (in the Sudanian zone) are the optimal rates that have induced optimal yield of 2.1 t⋅ha−1 (i.e. 2.5 times the yield in the farmers’ field) pod yield and the best return on investment per hectare. Nevertheless, for a sustainable groundnut producproduction, treatment 13.1-25.07-11.47-20-1.82 is suggested as regular K input is required for the respect of the fertilization laws.