logoalt Hacker News

khalictoday at 3:32 PM4 repliesview on HN

LPDDR6... that's not enough for speedy inference


Replies

hoherdtoday at 3:44 PM

Oh thank heavens, computing hardware that the AI companies will not buy all of the supply of.

CharlesWtoday at 3:43 PM

It depends on the number of channels. The Xring O3 appears to have 4×24-bit channels, so 113.8 GB/s. The iPhone 17 Pro's memory bandwidth is ~76.8 GB/s. Seems fine?

show 2 replies
simlevesquetoday at 3:40 PM

Not every computing platform is expected to do fast inference.

rjzzleeptoday at 3:38 PM

Maybe,but memory bandwidth is much better than on the DGX Spark at least.

show 1 reply