The energy density per se depends mostly on the electrode materials used. This battery uses conventional electrode materials, so in principle it can achieve the usual energy density of Li-ion cells.
The cells use Li4TI5O12 at the negative electrode, which means that they most likely operate at a lower nominal voltage than traditional Li-ion cells (somewhere around 3V as opposed to >4V), however since the electrodes can be made much more dense (because the liquid electrolyte does not need to penetrate in the electrode and therefore the empty porous space in the electrode is not necessary), overall the energy density of the electrode (capacity times voltage) remains the same.
The main advantage of solid state cells comes from the fact that the electrolyte layer, in principle, can be made much thinner than the usual ~20 microns required by the traditional polymeric separators. However, manufacturing such thin layers of purely ceramic materials is not easy, and I don't think (cannot access the paper right now) that the cell shown in the paper has any advancement in that respect. I think the best example of thin, ceramic, Li conducting layers are those made by Ohara in Japan.
So overall an interesting technological step, but no breakthrough.