OpenAI improves GPT-6 prompt caching with up to 90% input token savings

OpenAI upgraded GPT-6 prompt caching in its API, cutting input token costs by up to 90% and making agents faster. A new Prompt Caching Dashboard plus explicit cache breakpoints and context prewarming let developers track and maximize cache hits.

2026-09-23 ~ 2026-09-23 · 3 related posts

• OpenAI Boosts GPT-6 Prompt Caching, Input Tokens Now Up to 90% Cheaper — OpenAIDevs · 2026-09-23
• OpenAI Launches Prompt Caching Dashboard to Track Cache-Hit Rates and Misses — OpenAIDevs · 2026-09-23

1 near-duplicate retellings: OpenAIDevs