IPv6 unicast addresses can be aggregated with prefixes of arbitrary bit length, similar to IPv4 addresses under CIDR.

There are several types of unicast addresses in IPv6, in particular, global unicast, site‑local unicast (deprecated), and link-local unicast. There are also special‑purpose subtypes of global unicast, such as IPv6 addresses with embedded IPv4 addresses.

IPv6 nodes can have considerable or no knowledge of the internal structure of the IPv6 address, depending on the role the node plays (for instance, host versus router). At a minimum, a node can assume that unicast addresses (including its own) have no internal structure (as shown in Unicast Address with no Internal Structure).

A slightly sophisticated host (but still rather simple) can also be aware of a subnet prefix for the link or links to which it is attached. The different addresses can have different values for n (as shown in Unicast Address with Subnet Prefix).

Though a very simple router might have no knowledge of the internal structure of IPv6 unicast addresses, routers typically have knowledge of one or more of the hierarchical boundaries for the operation of routing protocols. The known boundaries differ from router to router, depending on the positions the router holds in the routing hierarchy.

Except for the knowledge of the subnet boundary discussed in the previous paragraphs, nodes should not make any assumptions about the structure of an IPv6 address.