Running inference for a model, even when you have all the weights, is not trivial.
Running inference for a model, even when you have all the weights, is not trivial.
Let's not even talk about the case when you have monitors that have different DPI, something that is handled seamlessly by MacOS, unlike Linux where it feels like a d20 roll depending on your distro.
I expect most desktop MacOS users to have a HiDPI screen in 2026 (it's just...better), so going to Linux may feel like a serious downgrade, or at least a waste of time if you want to get every config "right". I wish it was differently, honestly - the rest of the OS is great, and the diversity between distros is refreshing.
Linux has bugs, bug MacOS does too. I feel like for a dev like me, the linux setup is more comfortable.
* [1] https://sql-page.com/
The US has enough leverage over Germany that the Nord Stream incident is still an unsolved mystery where everyone just moved on. Probably the Russians did it. That's a lot of leverage.
You can see that 17% of answers come from India alone and that software developers got below average results, for instance.