Google TPUs, Amazon, a YC funded ASIC/FPGA company, a Chinese Co. all have custom hardware too that might scale well.

How exactly does groq scale wide well? Last I heard it was 9 racks!! to run llama-2 70b

Which is why they throttle your requests

Well, Cerebras pretty much needs a data center to simply fit the 405B model for inference.

I guess this just shows the insanity of venture led AI hardware hype and shady startup messaging practices