Ross 308 broilers in a randomized complete block design with a 2 × 2 factorial treatment arrangement (four treatments with 12 replications of six chicks each) were fed corn and SBMbased diets with two concentrations of metabolizable energy (ME) (normal (positive control, PC) and low (negative control, NC)) and two amounts of enzyme cocktail (EC) (0% and 0.005%) for 35 days. Performance, carcass traits, serum metabolites, ileal histology, and apparent nutrient digestibility were evaluated. Compared with the non-supplemented diet, the use of EC improved feed conversion ratio (FCR) over 26−35 and 0−35 days (p < 0.01), European performance efficiency factor (EPEF) over 26−35 days (p < 0.05), dressing yield (p < 0.01), villus height (p < 0.05), nitrogen-corrected apparent ME (AMEn) (p < 0.01), and serum glucose (p < 0.05). Compared with the NC diet, feeding the PC diet improved FCR over all experimental periods (p < 0.01, p < 0.05, p < 0.05, and p < 0.01, respectively), EPEF over 0−10 days (p < 0.05), and AMEn retention (p < 0.01). To conclude, the AMEn of broilers fed corn and SBM diets could be improved by adequately adjusting dietary ME and using a cocktail of non-starch polysaccharide-degrading enzymes, improving commercial benefits to producers.