Extend GPT-5.1 prompt cache retention to 24 hours
Use prompt_cache_retention='24h' for faster, cheaper follow-up questions.
Set to '24h'
API call parameter `prompt_cache_retention`
Allows prompts to remain active in the cache for up to 24 hours, reducing latency and cost for subsequent related requests in long-running interactions.
Extend prompt caching in GPT-5.1 to 24 hours using the `prompt_cache_retention='24h'` parameter. This significantly reduces latency and cost for follow-up questions in multi-turn conversations, coding sessions, or knowledge retrieval.
A longer cache retention means more follow-up requests can reuse previously processed information, avoiding redundant computation and network calls, leading to faster and cheaper responses.
For single-turn requests or when the context of follow-up questions is unlikely to overlap with previous ones.
- Introducing GPT-5.1 for developers ↗OpenAI · OpenAI News · Nov 12, 2025 · paraphrased
First seen on Subvolts Nov 12, 2025. Updated Sep 11, 2026. Spotted a problem? Tell us.
- Introducing GPT-5.1 for developersOpenAI · OpenAI News · Nov 12, 2025
The Playbook: reusable how-tos, prompts, configs and tips pulled out of the news.
- Cost tipGPT-5.1 prompt caching reduces costExtended prompt caching for GPT-5.1 makes follow-up questions cheaper by reusing context. · 2 min
- Code snippetGPT-5 custom tool calling with plaintext and grammarsEnable GPT-5 to call tools using plaintext and constrain with developer-supplied context-free grammars. · 5 min
- HeuristicDynamically adjust GPT-5.1 reasoning effort for speed and costBalance speed, cost, and intelligence by choosing the right `reasoning_effort` level for your GPT-5.1 task. · 5 min
- Setting tipGPT-5.1 no reasoning mode for faster responsesSet reasoning_effort to 'none' for latency-sensitive use cases with GPT-5.1. · 1 min
- ToolGPT-5.1 shell tool for running shell commandsEnables GPT-5.1 to execute shell commands on the user's local machine. · 5 min
- Model pickGPT-5 sizes for performance, cost, and latency trade-offsChoose from gpt-5, gpt-5-mini, and gpt-5-nano for flexible performance, cost, and latency. · 5 min
Questions people ask
- Which AI tools does this work with?
- The sources used it with ChatGPT.
- Where did this come from?
- OpenAI (OpenAI News). Subvolts collected, normalized and credited it; the link above opens the original.