Introducing l: A New Runtime for k and q with Modern Performance

A new runtime for k and q: l

I present l, a new runtime for the k family that executes k4, q, and qSQL unchanged on a modern engine. By keeping data compressed in memory and utilizing transparent SIMD and automatic parallelism, l moves fewer bytes to achieve massive speedups. You can run your existing code without rewrites while benefiting from optimized execution on Apple Silicon, x86, and GPU hardware.

"Modern hardware starves for bandwidth long before it runs short of compute — so moving fewer bytes is what makes it fast."