Hands-on: OpenAI's Astra 6.2 disappoints — writing regressed, coding stops too early

danshipper · x · 2026-11-18

Every's weeks-long testing of OpenAI's newly launched Astra 6.2 and Sol 6.2 finds the release disappointing versus Fable 5.2.

Retweeter TylerNishida: "this is the first time i felt that gemini is ahead of oai."

Related event: Every's Tests Find OpenAI's Astra 6.2 Disappointing in Writing and Coding(2 posts)→

Original post →

More from Models

Models channel →