What is the size of a register?

The size of a register, also called its width or length, is defined by the number of bits it can hold at one time and is directly tied to the processor's architecture and word size. Historically, early processors had 4-bit or 8-bit registers, then 16-bit, and for a long time 32-bit was standard. Today, most general-purpose processors are 64-bit architectures, so their main general-purpose registers are 64 bits or 8 bytes wide. However, there is variation within the same CPU. General purpose registers match the word size, but other registers have different sizes: the Program Counter might be 64-bit, status or flag registers might be 16-bit or 32-bit where only a few bits are used, segment registers in x86 were 16-bit, and floating-point registers are 80-bit in x87, while vector registers like SSE are 128-bit, AVX are 256-bit, and AVX-512 are 512-bit. Smaller microcontrollers still use 8-bit or 16-bit registers. So when asked about size, you must specify which register type and which architecture, but for a modern desktop CPU, you can assume a general-purpose register is 64 bits wide and can store values from 0 to 18 quintillion.