The generalized relativistic pseudopotential (GRPP) method is one of the most accurate approaches to account for relativistic effects in high-precision modeling of electronic structure and properties of atoms, molecules and materials. Here we present the latest developments in the LIBGRPP library for the evaluation of molecular integrals of the GRPP operator over Gaussian basis functions. The approach to the parallelization based on the OpenMP technology was proposed and implemented to make the library applicable in quantum chemistry simulations of large systems containing heavy elements on modern high-performance supercomputers. In addition, two Python modules, the first one providing an access to the core LIBGRPP routines written in C, and the second one allowing the interfacing with the PySCF program package, were developed. A parallel efficiency of the LIBGRPP library was demonstrated on a cluster model of the ytterbium trifluoride (YbF \(_3\) ) crystal, the 23.2x speedup in the calculation of GRPP integrals was achieved on 14 cores of the Intel(R) Xeon(R) E5-2680 v4 CPU.