Covers how to change effort mid-conversation without invalidating prompt cachingReusing the model's processed form of a repeated prompt prefix so subsequent calls skip re-reading it, cutting cost and latency substantially.Full definition →, plus cost and migration practices. Also explains that Mythos and Fable share weights with different safeguards, which affects access planning.
Terms in this piece · Glossary
prompt caching — Reusing the model's processed form of a repeated prompt prefix so subsequent calls skip re-reading it, cutting cost and latency substantially.
