Continuous heavy metal biosorption in membrane reactor apparatus using binary metal solutions was here considered. A dynamic model was developed to simulate biosorption performances on the base of metal mass balances in the system considering biosorption as an equilibrium process. The effect of three Langmuir-type competitive models on dynamic simulations of biosorption was then studied by using predictive equilibrium models (whose adjustable parameters were determined only by single metal system biosorption data) and not predictive equilibrium models (adjustable parameters directly by binary biosorption data). Predictive competitive models can give simulation profiles that are different from those obtained using non-predictive models. This detachment is due to the non-ideal competition among metals in solution which cannot be predicted only on the base of biosorption data in single metal systems. The dynamic model for multi-component biosorption here proposed was compared with experimental results reported in the literature and obtained using a biomass in a similar membrane reactor apparatus with ternary metal systems. The simulated profiles (obtained by using predictive equilibrium models) can reproduce qualitatively the specific adsorbent selectivity and the overshoot regions in the permeate concentrations of the metals with the minor affinity.