diff --git a/examples/queryGpu.f90 b/examples/queryGpu.f90
new file mode 100644
index 0000000..e14c4b3
--- /dev/null
+++ b/examples/queryGpu.f90
@@ -0,0 +1,29 @@
+program queryGpu
+  use cudafor
+
+  implicit none
+
+  real, parameter :: GBUnit = 1024 * 1024 * 1024.0
+  real, parameter :: GHzUnit = 1e6
+  integer :: istat, devId
+  type(cudadeviceprop) :: properties
+
+  ! Select a device to query
+  devId = 0
+  ! Query the properties of device devId
+  istat = cudaGetDeviceProperties(properties, devId)
+
+  write(*,'(A,I0,2A)') "Device ", devId, ": ", properties%name
+  write(*,'(A,T39,A,I0,A,I0)') "Compute capability", ": ", properties%major, ".",  properties%minor
+  write(*,'(A,T39,A,F5.2,A)') "Total amount of global memory", ": ", properties%totalGlobalMem / GBUnit, " GB"
+  write(*,'(A,T39,A,F4.2,A)') "GPU clock rate", ": ", properties%clockRate / GHzUnit, " GHz"
+  write(*,'(A,T39,A,F4.2,A)') "Memory clock rate", ": ", properties%memoryClockRate / GHzUnit, " GHz"
+  write(*,'(A,T39,A,I0)') "Number of SMs", ": ", properties%multiProcessorCount
+  write(*,'(A,T39,A,I0)') "Warp size", ": ", properties%warpSize
+  write(*,'(A,T39,A,I0)') "Maximum number of threads per SM", ": ", properties%maxThreadsPerMultiprocessor
+  write(*,'(A,T39,A,I0)') "Maximum number of threads per block", ": ", properties%maxThreadsPerBlock
+  write(*,'(A,T39,A,I0,A,I0,A,I0)') "Maximum sizes of each block dimension", ": ", properties%maxThreadsDim(1), &
+       " x ", properties%maxThreadsDim(2), " x ", properties%maxThreadsDim(3)
+  write(*,'(A,T39,A,I0,A,I0,A,I0)') "Maximum sizes of each grid dimension", ": ", properties%maxGridSize(1), &
+       " x ", properties%maxGridSize(2), " x ", properties%maxGridSize(3)
+end program queryGpu