We’ve improved prompt caching in the API for GPT-6, helping agents run faster and cost less. Higher cache-hit rates by default mean more input tokens benefit f

We’ve improved prompt caching in the API for GPT-6, helping agents run faster and cost less. Higher cache-hit rates by default mean more input tokens benefit from cached-input discounts of up to 90%. [▻](https://video.twimg.com/amplify_video/2102461494151852032/vid/avc1/1080x1080/VNPlQnjDxKhiakl…