IPv4 - classless addressing

IPv4's initial classful design proved to be less scalable in the face of the protocol's increasing usage, as many millions of addresses were wasted especially within class A subnets.

For this reason, classless subnetting was introduced. This is where the Variable Length Subnet Mask or VLSM] comes in. This essentially means that the classful "rule" that states that class A addresses can only have a class A subnet is relaxed. You are allowed to further subnet a class A address with smaller subnet masks.

This has allowed a more efficient usage of the IPv4 address space. That's why you see class A address spaces with subnet masks other than /8. In a classless network environment, the terms Class A, Class B, and so on are now used only as a reference and not as a concept that constrains the subnet mask size.

Links

https://networklessons.com/subnetting/variable-length-subnet-mask-vlsm