Biosurfactants, surface-active agents produced by microorganisms, are increasingly studied for their potential use in soil remediation processes because they are more environmentally friendly than their chemically produced homologues. In this work, we report on the use of a crude biosurfactant produced by a bacterial consortium isolated from a PAHs-contaminated soil, compared with other (bio)surfactants (Tween80, Sodium dodecyl sulfate - SDS, rhamnolipids mix), to wash PAHs from a contaminated porous media. Assays were done using columns filled with sand or sand-clay mixtures (95:5) spiked with four model PAHs. The crude biosurfactant showed less adsorption to the [sand] and the [sand + clay] columns compared to Tween 80, SDS and the rhamnolipid mix. The biosurfactant showed the second best capacity to remove PAHs from the columns (as dissolved and particulate phases), both from [sand] and [sand + clay], after SDS when applied at lower concentrations than the other sufactants. The effluent concentrations of phenanthrene (PHE), pyrene (PYR) and benzo[a]pyrene (BAP) increased in the presence of the crude biosurfactant. Compared to the control experiment using only water, the global PAHs washed mass (amount of PAHs removed from the columns) increased between 9 and 1000 times for PHE and BAP in the [sand] column, and between 55 and 6000 times respectively for PHE and BAP in the [sand + clay] columns. Moreover, in the [sand + clay] columns, leaching of a part of the clays was observed in the SDS and the biosurfactant injections assays. This clay leaching resulted in higher PAHs removal, due not to desorption but rather to particulate transport. In the context of washing PAH-contaminated soils in biopiles or subsurface remediation, our results could help in sizing the remediation approach using an environmental friendly biosurfactant, before a pump-and-treat process.