Question

A support team sends requests to Claude that all begin with the same 8,000-token knowledge base document, followed by a short, unique customer question that changes with every request. They enable prompt caching to reduce cost and latency. To get the maximum benefit from caching, where should the reusable content be positioned relative to the variable content in each request?

Accepted Answer

The stable knowledge base should come first, with the cache breakpoint placed after it, and the unique customer question placed after the cached portion.

Answer

The unique customer question should come first so it is cached, followed by the knowledge base document.

Answer

The knowledge base and the customer question should be interleaved so each cache entry stays small.

Answer

Caching only works when the entire request, including the customer question, is identical across calls, so ordering does not matter.