As the hardware continues to iterate at a rapid pace, anything you pick up second-hand will still deprecate at that pace, making any real investment in hardware unjustifiable.
Coupled with the dramatically inferior performance of the weights you would be running in a local environment, it's just not worth it.
I expect this will change in the future, and am excited to invest in a local inference stack when the weights become available. Until then, you're idling a relatively expensive, rapidly depreciating asset.
nowadays it's fancy touch display, requires concentration and often sluggish, and the machine often felt cheap and made cheap sound when tapped on, I don't think the operator are ever enjoying interacting with it and the software's often slow across the network....
I'm all for fast. It shows no matter what, at least somebody cared enough for it to be blazing fast.