Note what the article author writes: "On my machine the code runs at around 100 Gflops, which is not far from the theoretical peak, and similar to Intel's opti

Note what the article author writes: "On my machine the code runs at around 100 Gflops, which is not far from the theoretical peak, and similar to Intel's optimized implementation." So he knows that, for example, Intel provides/sells already optimized implementation. And he knows that there is GotoBLAS (1), because as you noted, he cites the paper about it. But he explains the implementation present in Glow, "a machine learning compiler and execution engine for various hardware targets." (2) Now if you want to prove to the authors of Glow why they should have "just used BLAS" I would like to read this discussion (or even better, see a proof of concept code with benchmarks etc). But I can imagine that in their case there were some reasons why not to use it, as they were aware that it exists. However it is true that the readers should be aware of the background assumptions and to be aware of the existence of the libraries. 1) https://en.wikipedia.org/wiki/GotoBLAS https://en.wikipedia.org/wiki/GotoBLAS 2) https://github.com/pytorch/glow https://github.com/pytorch/glow