I've run it on an LXC with 20GBs of ROM and 2GB of RAM, lol

True. People are buying them with the wrong specs / intentions. A $2600 Mac mini will actually run decent Qwen models on board and can mitigate token costs with a good routing policy, and/or an operator who knows when to delegate to cloud models. What seems expensive up front can actually be a good ...

I low keys still want it for iMessage though so I can show my friends the future Only $50/month

Anyone with experience running it on the new Umbrel ?