User reports Astra is faster and more token-efficient, asks OpenAI what changed

McDonaghMatthew · x · 2026-09-21

Developer McDonaghMatthew reports that Astra has become noticeably faster and even more token-efficient, directly asking OpenAI "what did you do?"

The post includes a comparison, suggesting OpenAI quietly shipped server-side optimizations or changes to Astra without announcement.

Original post →

More from Models

Models channel →