OpenAI quietly boosts some of Astra’s evaluation metrics, and continues to change others post-launch | Fortune0

• AI, bond-market, Business, Business / Artificial Intelligence, Business / Big Tech, Business / Companies, Business / Computers and Software, Business / Markets, Business / Social Media, business-idea, business-insider-today, business-strategy, CoECI News, E! Insider, Energy & Environment Newsletter, Finance, Gaming News Roundup, Gear / Gear News and Events, Global, Global Precipitation, global warming, globalstar, HEALTH, Home, House, insider-picks, insider-picks-guides, insider-reviews, insider-reviews-2021, insider-today, job-market, Markets, Markets & Festivals, News, newsletter, Newsletters, Nexstar Media Wire News, People News, Personal Finance, personal-finance, Photovogue / Editorial / News, Politics / Politics News, Property / News, Reviews, reviews-rit-ads, Security / Security News, shop, Shop Beauty, Tech
• September 5, 2026

OpenAI has changed several evaluation benchmarks for its GPT-6 Astra model since first publishing a blog post announcement mid-afternoon on Sept. 3. In some cases, the numbers on the updated versions showed Astra performing better, while numbers for models from OpenAI’s arch rival Anthropic got worse. The changes occurred amid an unusual rollout of the