Comment on Researchers upend AI status quo by eliminating matrix multiplication in LLMs

bitfucker@programming.dev ⁨1⁩ ⁨week⁩ ago

• pennomi@lemmy.world ⁨1⁩ ⁨week⁩ ago

  Only for maximum efficiency. LLMs already run tolerably well on normal CPUs and this technique would make it much more efficient there as well.

  source