Comment by jart

The inequality between matrix multiplication implementations is enormous. It gets even more extreme on GPU where I've seen the difference between naïve and cuBLAS going as high as 1000x. Possibly 10000x. I have a lot of faith in myself as an optimization person to be able to beat compilers. I can even beat MKL and hipBLAS if I focus on specific shapes in sizes. But trying to beat cuBLAS at anything makes me feel like Saddam Hussein when they pulled him out of that bunker.