AIOpenAI updates GPT-6 API prompt caching and adds a monitoring dashboard
OpenAI says higher cache-hit rates by default mean more input tokens benefit from cached-input discounts of up to 90%, helping agents run faster and cost less.
TLDR
OpenAI announced a GPT-6 API prompt-caching update that it says increases cache-hit rates by default, helping agents run faster and cost less. The company says its new Prompt Caching Dashboard lets developers track cache-hit rates, while its diagnostics API can identify changes that prevented reuse and estimate how many tokens were affected.
OpenAI updates GPT-6 API prompt caching and adds a monitoring dashboard
OpenAI says higher cache-hit rates by default mean more input tokens benefit from cached-input discounts of up to 90%, helping agents run faster and cost less.
TLDR
OpenAI announced a GPT-6 API prompt-caching update that it says increases cache-hit rates by default, helping agents run faster and cost less. The company says its new Prompt Caching Dashboard lets developers track cache-hit rates, while its diagnostics API can identify changes that prevented reuse and estimate how many tokens were affected.
