32GB Unified RAM max. Is that really the machine I want for running local LLMs?

Tagbert 6 hours ago | parent | next [-]

This M6 is the base Pro model. You probably want the M5 pro/max for your LLMs.

alluro2 4 hours ago | parent | prev | next [-]

Not really - it seems that M7 will come out in Q1 27 and be targeted at that use-case more specifically.

	▲	skohan 4 hours ago \| parent [-]
		Base M7 probably will still have pretty low memory bandwidth for serious AI workflows. M7 max/ultra might be pretty interesting though.

Nope.