PRODUCTION SCENARIO
Ridgeway Compliance runs a gpt-5.6-sol review job every two hours. Each run sends 30 requests that share a 6,000-token instruction block and finishes in about ten minutes. Usage shows cached_tokens on every request after the first in a run, and never on the first.
Why does the first request of every run miss the cache?
Answering here is anonymous. Nothing is saved unless you sign in.
Show answer and explanation
Answer: A cached prefix expires 30 minutes after its most recent write or reuse by any request
On GPT-5.6 and later a cached prefix stays eligible for reuse for 30 minutes after its most recent write or reuse, so the two-hour gap between runs lets the entry lapse and the first request writes it again. The 6,000-token block clears the 1,024-token minimum, caching is on by default so no prompt_cache_key is needed, and prefix caching is not scoped to a previous_response_id chain.
Official Source: https://developers.openai.com/api/docs/guides/prompt-caching ↗