At its ambitious "Ampere" A100 Tensor Core processor reveal, the "Tesla" brand was nowhere to be seen. Heise.de reports that the company has decided to discontinue "Tesla" as the top-level brand for its HPC, AI, and scalar compute accelerator product line. NVIDIA introduced the Tesla compute accelerator brand along with its "Tesla" graphics architecture. It was the first major GPGPU product, and saw CUDA take flight as a prominent scalar compute language.

Over the years, NVIDIA kept the Tesla moniker as a top-level brand (alongside GeForce and Quadro), with an alphabetic portion of the model numbers denoting the graphics architecture the accelerator is based on (eg: Tesla P100 being "Pascal" based, K10 being "Kepler" based, and M40 being "Maxwell" based). The Tesla T4, based on "Turing," is the final product with the old nomenclature. Interestingly, Heise reports that NVIDIA dropped the name to avoid confusion with fellow Californian brand Tesla Inc.