1
Apple’s In-House Technologies Team Still Has Plenty of Work to Do (www.bloomberg.com)
submitted 2 years ago by [B] to c/apple@hardware.watch
33 comments fedilink hide all child comments
you are viewing a single comment's thread
view the rest of the comments
[–] [B] 1 point 2 years ago (2 children)

An obvious thing is if they want to beef up their neural engine to better handle AI stuff ? If they are going to add ‘on device’ LLM runtime support ?

  • permalink
  • fedilink
  • source
  • hideshow 2 child comments
    •
  • [–] [B] 1 point 2 years ago (1 child)

    LLM, emphasis on the Large. A standard LLM will take up A LOT of storage space to the point where the user won't have much left. Let alone the continuous processing power needed.

    Siri + local "Small" Language Model. It could have full access to on device data, but more generalized quires are done remotely.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 1 child comment
    •