I don’t think it is a nvidia driver issue but this will let us figure it out. My own personal experience aside (running 5.7 kernel with the 440x nvidia drivers with no memory leaks at all, leave my system running 24/7, play games, have 8 docker containers running various software like my media server, typically have 30+ tabs open in firefox and chromium at the same time, My current uptime on my system is 4days 14hours and my memory is sitting right around 10gb used of 32gb. That is normally where it sits. The only reason my uptime is only 4 days 14 hours is because that is when I rebooted my system to install the new update. Prior to that I was running 100% without a reboot since the update prior to that one.
All of that aside if there were an actual memory leak in the nvidia drivers that is something that people catch onto pretty quickly and the 440x drivers have been out for quite some time now.
So for troubleshooting, After freshly booting your system run the following command (this will list all processes sorted by memory usage and dump it into a file called process.log, you may call the file whatever you want.):
top -b -o+%MEM -n1 >process.log
Then launch your normal workflow/apps and run the command again
top -b -o+%MEM -n1 >process2.log
When you notice the problem with running out of memory then run the command for the 3rd time (note that you can feel free to take a snapshot of your current processes more often if you’d like. then if you post those files or paste the contents here into separate pre-formatted text blocks we can probably figure out what is eating up your memory.
top -b -o+%MEM -n1 >process3.log
(paste them into a message like you did your configuration so it will be easy to read and analyze).