OpenAI's /v1/responses API now supports an optional 12-hour prompt caching duration, keeping shared instructions, tools, and context cached longer so requests hours apart still get lower cached-input pricing. Cache writes for the 12-hour option cost 2.5x the input token price.
Read the full story on developers.openai.com
What to doConsider enabling 12-hour prompt caching for workloads with infrequent repeated requests, and factor the 2.5x cache-write cost into pricing.
- Provider
- OpenAI
- Source
- developers.openai.com