ABSTRACT Fish by-products were fermented with Lactobacillus plantarum at 30ºC using molasses as carbohydrate source. The ensilage process was completed after 30 days and at the end, a desirable and stable pH (4.5) was attained. Dried tomato by-product meal (TBM) and potato by-product meal (PBM) were used as alternative filler and blended with the liquid silage (40:30:30 w/w) and sun-dried. The resulting dried silage meal was included in the experimental diets to replace 0, 10, 20, 30, 40 or 50% of dietary protein (by weight) in isonitrogenous (30% CP) and isocaloric (2700 kcal ME/kg) pelleted diets. No significant differences were found for apparent digestibility coefficient (ADC) of dry matter (DM), crude protein (CP), ether extract (EE) and nitrogen free extract (NFE) up to 30% inclusion level, while the highest inclusion levels (40 or 50%) significantly reduced ADC for DM, CP, EE and NFE. In a 90-day growth trial, the replacing of dietary protein by fish silage protein up to 30% of dietary protein in tilapia diet had no significant effect on growth performance including final body weight (BW), body length (BL), condition factor (K), weight gain (WG) and specific growth rate (SGR). The highest replacing levels (40 or 50%) significantly (P