#prompt caching
4 articles tagged prompt caching. At twelve articles this topic starts lifting the whole site in search.
Timeline
-
Why AI API prices fell up to 80% in four months — cache reads are 96–99% of agent spending
-
Anthropic's Claude Fable 5.1 — same sticker price, cache reads cut 75%
-
What prompt caching is — from the second call, your input costs one-fortieth
-
What AI token pricing is — why output costs about 5× input
All articles
-
Tech · 4 min readWhy AI API prices fell up to 80% in four months — cache reads are 96–99% of agent spending
Cuts — GPT-5.6 Luna −80% on Jul 30; Opus 5.5 40–60% cheaper in Sept; GPT-6.1 Sol cache halved Sept 29
-
Tech · 4 min readAnthropic's Claude Fable 5.1 — same sticker price, cache reads cut 75%
Price — input $10 and output $50 unchanged; cache reads $1.00 → $0.25 (−75%)
-
Tech · 4 min readWhat prompt caching is — from the second call, your input costs one-fortieth
Mechanism — an identical prefix yields identical intermediate values, which can be stored
-
Tech · 5 min readWhat AI token pricing is — why output costs about 5× input
Billing is per token, with separate input and output rates — output typically costs four to five times input