Getting 50 GB/S Back from the Apple Neural Engine

Eileen YoonHome  |  About  Getting 50 GB/s Back Out of the ANEAug 10, 2026(3162 words)Introduction An RTL performance erratum in the Apple M3 Neural Engine throttles DRAM weight streaming throughput down to 17–19 GB/s from the nominal 45–60 GB/s, whenever the total weight size is an integer multiple of 1 MiB, which currently affects 7 of ANEMLL’s 15 models. Avoiding the problematic path in the kernel DMA engine's speculative prefetch ring increased Llama 3.2 1B toke