Left panel: residual multiplicative bias (top) and additive bias (bottom) as a function of r_sep, the distance to the nearest galaxy in the input catalogue brighter than m = 26. Right panel: residual biases as a function of n_fac, the relative increase in galaxy density, where n_fac corresponds to the grid-based case. In both cases, results were obtained using b^sm = 1.217, the best estimate for the fiducial setup. The results indicate that the performance is not affected significantly by blending, although the additive biases exceed Euclid requirements for extreme over- and under-densities.