Eight iso-nitrogenous and iso-energetic (crude protein, 48% and gross energy, 18.5 MJ/kg) diets (L5, L10, Z5, Z10, X5, X10, M5, and M10) were formulated with 5% and 10% waxy corn starch (L), high-amylose corn starch (Z), wheat starch (X), and cassava starch (M) to determine a suitable starch source and supplementation level for a largemouth bass, , diet. Fish (initial body weight, 23.46 g±0.19 g) were fed to apparent satiation twice daily for 45 d. The results showed that the supplemented dietary starch level and starch source had significant effects on growth, feed utilization, body composition, and non-specific immunological indices of largemouth bass. Significant decreases in specific growth rate and feed intake were observed with increasing content of the same starch from 5% to 10%. However, feed and protein efficiency increased significantly in the L10, X10, and M10 groups, compared with those in the L5, X5 and M5 groups, whereas no differences were detected in these parameters between the Z5 and Z10 groups. Hepatosomatic index, viscerosomatic index, and liver glycogen concentration increased significantly in the L10, X10, and M10 groups, compared with those in the L5, X5, and M5 groups, but no difference was found in the viscerosomatic index between the Z5 and Z10 groups Whole-body and liver lipid contents, as well as liver protein content decreased significantly when any of the supplemented starches were increased from 5% to 10% in the diets. Blood erythrocyte count and hematocrit in the L10 group, serum triglyceride content in the Z10 group, and hematocrit and serum triglyceride content in the X10 group decreased significantly, compared with those in the L5, Z5, and X5 groups. Serum 50% hemolytic complement (CH₅₀) activity was significantly lower in the Z10 group than that in the Z5 group. Serum glucose concentrations from 3 to 12 h after a meal in the L5, Z5, X5, and M5 groups were lower than those in the L10, Z10, X10, and M10 groups. Apparent digestibility of the dietary starch and hepatosomatic index were significantly lower in the Z5 group compared with those in the L5, X5, and M5 groups, whereas whole-body lipid deposition rate increased significantly. Apparent digestibility of dietary protein was significantly lower in the M5 group than that in the L5, Z5, and X5 groups. Apparent digestibility of dietary starch, hepatosomatic index, viscerosomatic index, and liver glycogen concentration decreased significantly in the Z10 group compared with those in the L10, X10, and M10 groups, whereas whole-body lipid deposition rate increased significantly. Apparent digestibility of dietary protein was significantly lower in the M10 group than that in the L10, Z10, and X10 groups. Significant interactions were detected between feeding rate, specific growth rate, protein digestion rate, lipid deposition rate, hepatosomatic index, glycogen content, erythrocyte count, serum triglyceride level, and serum CH₅₀ activity between the dietary starch supplementation level and starch source. In conclusion, the starch source and supplementation level in the largemouth bass diet exerted different effects on growth performance, body composition, hematological parameters, non-specific immunological indices, and serum glucose. Adding 10% high-amylose corn starch was the most appropriate supplement in the diet for largemouth bass.