CPU vs GPU vs TPU

CPU is the processing unit that works as the brains of a computer designed to be ideal for general-purpose programming.
GPU is a performance accelerator that enhances computer graphics and AI workloads.
TPUs are Google's custom-developed processors that accelerate machine learning workloads using (specific machine learning framework) TensorFlow
Central Processing Unit (CPU)
A Central Processing Unit (CPU) is the core processor designed with a few powerful cores and large
cache memory that enables it to run a few software threads at once.
CPU Features Summary: Has Several Cores, Low Latency, Specialized in Serial Processing, Capable of executing a handful of operations at once
Graphics Processing Unit (GPU)
A GPU (graphics processing unit) is a specialized processor that works as a performance accelerator with the CPU. GPU has thousands of cores that can break down complex problems into thousands or millions of separate tasks and work them out in parallel.
GPU Features Summary: Has thousands of cores, High throughput, Specialized for parallel processing, Capable of executing thousands of operations at once
Tensor Processing Unit (TPU)
A tensor processing unit (TPU) is an application-specific integrated circuit (ASIC) specifically designed to accelerate high-volume mathematical and logical processing tasks typically involved with machine learning (ML) workloads.
TPUs Features Summary: Special Hardware for Matrix Processing, High Latency. Very High Throughput, Compute with Extreme Parallelism