[otabdeveloper4]

Using a full-blown GPU just for neural network inference is crazy inefficient. They should hire some blockchain dudes to build them custom hardware for one tenth of the price.

[otterley]

NPUs already exist and are everywhere (Apple Silicon, AWS Inferentia). AWS is leading at scale in the cloud in this space.

(Disclaimer: I work for AWS, but opinions are my own.)

[otterley]

I failed to include Google TPUs. My apologies, GCP folks!

[fragmede]

Google hired a bunch of chip designers to make TPUs for their ML people, so are highly optimized for this kind of work, which are currently on their 5th gen, and are available on their cloud.

(Disclaimer: I used to work there but not on them.)