A 12-week feeding trial was conducted to evaluate the optimum dietary inorganic copper (copper sulphate) in juvenile beluga, Huso huso. Eight semi-purified diets containing 1.1 (Cu1.0), 3.5 (Cu4.0), 7.1 (Cu7.0), 9.7 (Cu10), 13.1 (Cu13), 25.1 (Cu25), 49.9 (Cu50) and 195 (Cu195) mg Cu kg−1 diet in the form of CuSO4.5H2O were fed to fish of initial body weight 8.49 ± 0.32 g and length 11.85 ± 0.66 cm (mean ± SD) in triplicate groups in a flow-through system. Weight gain (WG) of fish fed Cu10 and Cu13 diets was significantly higher than that of fish fed Cu1.0, Cu4.0, Cu25, Cu50 and Cu195 diets (P < 0.05). Whole-body and muscle crude protein increased with dietary Cu up to the supplementation level of 13.1 mg kg−1 diet and then decreased. Whole-body lipid content was negatively correlated, while whole-body ash was positively correlated with dietary copper concentration. Hepatic copper–zinc superoxide dismutase activity of fish fed Cu10 and Cu13 diets was significantly higher than that of fish fed Cu1.0, Cu4.0 and Cu195 diets. Hepatic thiobarbituric acid-reactive substances of fish fed Cu13 diet was significantly lower than those of fish fed the other diets except for that of fish fed Cu10 diet. Aspartate aminotransferase, alanine aminotransferase and copper accumulation in tissues increased with dietary copper. Broken-line analysis of WG suggested that the optimum dietary Cu level was 10.3 mg Cu kg−1 diet. Therefore, these results may indicate that the optimum dietary Cu levels could be greater than 10.3 mg Cu kg−1 diet but less than 13.1 mg Cu kg−1 diet in juvenile beluga, when copper sulphate is used as the dietary source of inorganic copper.