What is your typical memory/cpu usage per instance?

I have theia deployed in kubernetes (not che), and I’ve noticed it takes up quite a bit of memory per instance. See output of kubectl top pods below for 5 open projects:

NAME                                                              CPU(cores)   MEMORY(bytes)
theia-deployment--am091610--john-74d5648bd6-hns5c                 122m         334Mi
theia-deployment--db100808--bookstore-b7bbf46c8-znrh2             119m         249Mi
theia-deployment--rc090209--ice-cream-you-scream-c64c988487nshn   258m         577Mi
theia-deployment--robotben--test-project-6fcfbc988b-ljghz         115m         492Mi
theia-deployment--trenzor--ice-cream-you-scream-6d8bd954f6jgw65   153m         353Mi

Is this to be expected? I’m curious what other users’ experience is. What is your mem/cpu usage on average? What factors tend to increase resources? Do you have any tips on reducing the memory footprint?