Re:Microsoft BitNet: 100B Param 1-Bit model for local CPUs
by bigyabai on Wednesday March 11, 2026 @01:15PM
(#47338349)
For larger contexts, the bottleneck is probably token prefill instead of memory bandwidth. Supposedly prefill is faster on the M5+ GPUs, but still a big hurdle for pre-M5 chips.
[ Parent
| on HN ]
There are no comments on this comment yet.