How to dedicate all my GPU memory to CUDA and not using it for display output?

I have an Intel CPU with iGPU and also a dGPU (Nvidia). I intend to use the dGPU for its CUDA so as the result, my monitor is connected to the iGPU.

Still when I open the nvtop tool, I can see that some of the dGPU memory is consumed even though I’m not doing anything with it. I assume it’s because the dGPU is still part of the display output even though the cable is connected to the motherboard. I have this suspicion especially since if I take the HDMI out and connect it to the dGPU, I’ll get video output.

My question is, how can I ask the OS not to use the dGPU as part of the display output and leave all its memory unallocated?

My current display driver situation:

sudo mhwd -li
> Installed PCI configs:
--------------------------------------------------------------------------------
                  NAME               VERSION          FREEDRIVER           TYPE
--------------------------------------------------------------------------------
     video-modesetting            2020.01.13                true            PCI
           video-linux            2018.05.04                true            PCI
video-hybrid-intel-nvidia-prime            2021.11.04               false            PCI