A Graphics Processing Unit (GPU) is an electronic circuit designed for highly parallelized computations. GPUs are efficient in processing images, computer graphics and neural networks used in AI. In comparison to central processing units (CPUs) which only have a couple of cores, GPUs have hundreds or even thousands of cores enabling faster computation through parallel processing.
NVIDIA and AMD are the two major GPU manufacturers worldwide. NVIDIA through its CUDA library and AI optimized Tensor cores is generally considered having an advantage compared to AMD in terms of usability for AI computing.