Diagnosing foliar nutritional status is essential for fertilizer recommendations and for the identification of nutrient imbalances. This study aimed to verify genetic diversity and establish mean standards (leaf nutrient contents; LNCs) and relationships among leaf nutrients (LNC relationships; LNCRs) in seven conilon coffee genotypes during both pre-flowering and bean-filling stages. Twenty crops from several cities in the northern region of Espírito Santo State, Brazil, with crop yield either equal to or greater than 100 bags per hectare (during two harvests) were assessed. A total of 140 samples were collected during each evaluation period for quantification of leaf nutrient contents (N, P, K, Ca, Mg, S, Fe, Zn, Cu, Mn, and B). The Ward procedure, a hierarchical genetic clustering method, was used to quantify the genetic diversity among genotypes. To examine differences between the LNCs and LNCRs, F-and Scott-Knott tests were used. LNCs and LNCRs showed significant differences among the conilon coffee genotypes during the evaluation periods. Additionally, the 8V, 10V,and 12V genotypes exhibited the highest values for most of the nutrients, especially for N, P, and Cu. Therefore, the clustering method revealed genetic diversity among genotypes for standard leaf nutrient levels, implying that leaf diagnosis could be specific to each genotype and phenological stage.