I use a p4000 to transcode Plex content. The GPU uses less power than maxing out my CPU. I also have the GPU available in Kasm though I’m unsure if that works as I mostly use kasm as a SSH jumpbox or access to an unfiltered Chrome session at work.

When I upgrade my CPU, I plan to put my spare 1650 in my host and spin up a windows gaming VM for guests and remote play.

Just about anything Machine Learning or AI, Transcoding for a media server, render farm for something like blender perhaps?

Get an Intel cpu with iGPU (most do) and you’ll be good to go for anything that you’re doing there.

I have two GPUs in a single tower.

A GTX 750 to that I share with my LXCs. It does jellyfin transcode, frigate nvr for 3 cameras, kasm accelerated desktops, xfce4 pve host acceleration, Jupyter tensorflow, ersatz tv transcode, and I plan to use it for immich. At most it is taxed about 25 percent but I plan to have a lot more nvr and jellyfin streams.

I also have a 1660 ti passed to windows 11 VM for my gaming VM. I use sunshine and moonlight for remote gaming but I also roll easy diffusion for some image generation. I had an LLM but (https://github.com/oobabooga/text-generation-webui) but it was too slow for what I’m used to - I just use bing chat and now meta on whatsapp for my personal and an LLM I have access to at work.

If you run an AI server or crypto miner. Mostly

Yes, GPU for transcoding is reason enough, but object detection using frigate, or AI stuff is nice also. Buy a nvidia.