iPhone 17 Pro Demonstrated Running a 400B LLM

Re:iPhone 17 Pro Demonstrated Running a 400B LLM by fouc on Tuesday March 24, 2026 @07:41AM (#47501223)

looks like they meant “effective dense size” which is the square root of total params×active params, so in this case sqrt(397 x 17) = ~82

[ Parent | on HN ]

Re:iPhone 17 Pro Demonstrated Running a 400B LLM by zozbot234 on Tuesday March 24, 2026 @09:56AM (#47502689)

But the claim that "one expert is 17B" is incorrect. Experts are picked with per-layer granularity (expert 1 for layer X may well be entirely unrelated to expert 1 for layer Y), and the individual layer-experts are tiny. The writeup for the original experiment is very clear on this.

[ Parent | on HN ]

Re:iPhone 17 Pro Demonstrated Running a 400B LLM by stingraycharles on Tuesday March 24, 2026 @01:43PM (#47506431)

Ok I am by no means an expert on this and I immediately stand corrected. But as I understand it, in order to understand the amount of active memory that’s required, it’s more accurate to go by the ~82B number, right?

[ Parent | on HN ]

Re:iPhone 17 Pro Demonstrated Running a 400B LLM by zozbot234 on Tuesday March 24, 2026 @02:36PM (#47507173)

The ~82B figure is an attempt to compare performance to an equivalent dense model. The amount of active parameters is given by the ~17B.

[ Parent | on HN ]