Comment by khalic

7 hours ago

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

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

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?

  • According to Mark Gurman, the base M6 is bumped up to 200 GB/s of memory bandwidth with a single core performance bump of 15%.

  • If I recall correctly, Strix Halo gets about 450G/s. The M series processors get between 400 and 800G/s, and an actual NVIDIA card is like 5.5T/s. 113G/s is pretty slow for inference.

    • The CPU is for mobile phones. An Nvidia H200 has 4.8TB/s. An RTX 5090 has 1.8TB/s. Both use ~700W - not comparable with a phone.