logoalt Hacker News

StevenWatermanyesterday at 9:51 PM1 replyview on HN

Cached tokens are cheaper (90% discount ish) but not free


Replies

moyixyesterday at 10:11 PM

Also, unlike OpenAI, Anthropic's prompt caching is explicit (you set up to 4 cache "breakpoints"), meaning if you don't implement caching then you don't benefit from it.

show 1 reply