OpenAI's 12-hour prompt cache: cached-input pricing even when requests are hours apart
What is it?
In the OpenAI Responses API you can now pick a 12-hour minimum lifetime for prompt caches. On GPT-5.6 and later, set prompt_cache_options.ttl to "12h"; the default is "30m". You can mix both in one request, for example 12 hours for shared instructions and 30 minutes for recent turns.
Why does it matter?
Agents that run all day often go hours between requests, so their caches kept expiring. A 12-hour cache write costs 2.5x normal input (1.25x for 30 minutes), but reads cost 0.1x (0.05x on GPT-6.1 Sol). If long instructions and tool definitions get reused several times a day, you save. Put a 12-hour TTL on content you only reuse once or twice and you lose money.
Maturity
GAGenerally available in the Responses API. Chat Completions and models before GPT-5.6 keep using prompt_cache_retention.
How to track it
- Pricing: watch the API pricing page for changes to 12-hour cache-write rates per model.
- Scope: check the docs for the 12-hour option reaching Chat Completions or the Agents API.
- Measure: compare hit rates and cost before and after in the prompt caching dashboard.