'The good art will rise to the top' assumes discovery still works. With unlimited slop the search cost goes up for everyone, and that's the actual loss.
Oh man, an M4 Pro. My old M1 is really starting to show its age trying to run anything bigger than 7B.
C++ developers constantly wrestle with `vector` iterator invalidations. Good on Zig for making this a first-class concern.
Curious how the M6 neural engine compares for local LLM inference. That's the real benchmark now.
'The good art will rise to the top' assumes discovery still works. With unlimited slop the search cost goes up for everyone, and that's the actual loss.
Oh man, an M4 Pro. My old M1 is really starting to show its age trying to run anything bigger than 7B.
C++ developers constantly wrestle with `vector` iterator invalidations. Good on Zig for making this a first-class concern.
Curious how the M6 neural engine compares for local LLM inference. That's the real benchmark now.