Yes, if you run PyFR with a command something like mpiexec -n 4 pyfr run -b cuda ... that should do the trick. If you haven’t already it’s probably worth compiling MPI with the CUDA aware flag, see here for more details: Performance Tuning — Documentation