Experimental data concerning the bioavailability of the different Mg-salts in human organism is inconsistent. Mg-absorption reported by clinical studies largely varies depending on the method used for evaluation. The aim of this study was to evaluate the bioavailability and accessibility of magnesium bound in different Mg-salt compounds, using an in vitro model of intestinal cell barrier. The study included a variety of inorganic (oxide, sulphate, chloride, carbonate) and organic salts (lactate, citrate, pidolate). Caco-2 cells were cultivated in a complete culture medium with different magnesium salts treatments in ascending concentrations. The viability and quantity of cells was analysed by FACS. Mg-absorption was analysed by a direct colorimetric assay, measured by spectrometry. T-test identified a significant decrease in cell count treatment with mg-lactate compared with citrate. Mg-pidolate showed a significantly higher cell viability compared with Mg-citrate, Mg-lactate and Mg-chloride. Even though the difference was not significant, we showed that an increase in Mg2+ salt concentration progressively decreased the cell count and the viability and the effect was universal for all the used Mg-salt treatments. Mg-citrate, chloride, and sulphate showed a significantly lower absorption compared to Mg-carbonate, pidolate and oxide. Our in vitro monolayer model of human intestinal transport showed that viability and quantity of cell decreased with increasing Mg-concentration. We admit that our experiment model may have some limitations in accurately describing an in vivo Mg2+ absorption. Moreover, it is also necessary to assess the relevance of our data in vivo and especially in clinical practice.