X11 DISPLAY Configuration for SSH and Nvidia Optimus Graphics

There are one Intel integrated graphics and the other Nvidia graphics card in my remote box, where OpenSSH and X11 servers run on Ubuntu Desktop 24.04. The graphics drivers and utils are installed properly as well. $ lshw -c video *-display configuration: driver=nvidia *-display configuration: driver=i915 $ nvidia-smi -L GPU 0 $ prime-select Usage: /usr/bin/prime-select nvidia|intel|on-demand|query $ prime-select query on-demand I think that the term *on-demand* means my card enabled by Nvidia Optimus . However, it is not listed in the supported GPUs mentioned by the Debian Wiki . The list shows there're totally 3 pages, but it remains on page 1 full of old GPUs, no matter I click on next pages using the latest Chrome browser. My first question is whether the poorly maintained web page indicates that the *Optimus* is rather a default feature enabled for all the recent GPUs now, although Nvidia previously promoted it as a selling point for some of old GPUs. My second question is how to adjust the environmental variable *DISPLAY* for X11 server (preferably configured in the *~/.bashrc* file on the remote box) to follow the *on-demand* (ie. dynamically switched) graphics between the integrated and dedicated ones. My use case is that I *ssh* from the local terminal console to the remote box aforementioned, occasionally I need run X11 programs and display graphics on the remote box. Please note that I don't *ssh -X* (ie. nothing to do with X11 forwarding) for the local SSH client runs on a terminal console without X11.