ZORA relativistic DFT calculations are presented which aim to reproduce geometric structures and EPR properties of [Ni(mnt)(2)](-) (H(2)mnt = maleonitrildithiol), two other paramagnetic low-spin Ni(III) complexes, and an asymmetric paramagnetic Co(II) complex. The study tests the accuracy of the computational method as a prior step to the modeling of the geometric and electronic structure of the active site of NiFe hydrogenases in its EPR-active oxidized states Ni-A and Ni-B. Systematic deviations from experiment are found for the calculated g-values; relative differences among them are, however, well reproduced. Because no significant improvements have been achieved by using larger basis sets or more sophisticated functionals, g-values may be calculated rather rapidly at the VWN level. This is most important for the modeling of the active site of NiFe hydrogenases because its complexity does not permit calculations at high levels of theory. For [Ni(mnt)(2)](-), excellent agreement between calculated and experimental results is obtained for the N-14 quadrupole coupling, whereas the calculated hyperfine couplings are not always in good agreement with experimental data.