←Back to NewsAI News/LlmsnewsLlmsApiOpenAI Makes GPT-6 Prompt Caching 90% Cheaper and Fully VisibleOpenAI overhauls prompt caching for GPT-6 with a new dashboard, miss diagnostics, explicit breakpoints, and prewarming to cut agent latency and cost.SourceOpenAI DevelopersPublishedSep 22, 2026, 9:13 PMAuthorAlphaSignal NewsroomRead1 min readOpenAI overhauls prompt caching for GPT-6 with a new dashboard, miss diagnostics, explicit breakpoints, and prewarming to cut agent latency and cost.Reporting is indexed from AlphaSignal. Rights remain with the original publisher and cited sources.Read original report ↗Next readsOpenAI · newsOpenAI Drops GPT-6 Sol and Luna at 50% Less Than GPT-5.6SpaceXAI · newsxAI Ships Grok 4.7 With Longer Agent Runs at Unchanged PricesOpenAI · newsOpenAI's Astra for Law Pushes Into Big Firms With 54% Research Accuracy
