Under the hoodTier: how much it should matter to you
1Useful: my rating, 1 to 5
1Signal: worth watching, 1 to 5
Prompt CachingArea: what it touches
Internal ChangesKind: in v2.1.300,
Internal ChangesSection of the release
What
The prompt cache lets Anthropic's API reuse parts of a conversation it has already processed. Writing to it costs tokens, and entries can last 5 minutes or 1 hour.
Claude Code's record of each API request now includes cache_creation_5m_tokens and cache_creation_1h_tokens. Before, it only had one combined total.
Why
It shows how much cache cost comes from each cache lifetime.
How sure we are
One source agreesOne thing we can check says the same as this entry.
Anthropic's release notes agreeAdded cache_creation_5m_tokens and cache_creation_1h_tokens to the OpenTelemetry api_request event, splitting cache writes by lifetime
