I read the Hugging Face Kernels docs, then ran the kernels on my own CPU. RMSNorm was 3 to 9 times faster and usually bit-exact; rotary was slower. Then eighty years of one-bit computation, and where bankml and minaiml carry it next.

Dispatch #1: bankml goes live, Day 22 of the Book, a journal that shows a nearly full disk, and RAGE (2024) carried forward to 2026.

What 1-bit and ternary weights are, why the better ternary model ran seven times slower on x86, and how bankml, a zero-dependency Rust runtime, closed the gap at 9.5x per matrix with every bit proven against llama.cpp.