Comment by numbers_guy

Does anyone know why they brand it an "inference chip"? Is it something at the hardware level that makes is unsuitable for training, or is it simply that the toolchain for training is massively more complicated to program?