User Claims New GPT 6.2 Astra Got Worse at Making Slides
zck · x · 2026-11-18
A user compared OpenAI's own announcement post and claims GPT 6.2 Astra actually got worse at making slides. A personal observation without hard benchmarks, but part of the recurring debate over capability regressions after model upgrades.
More from Models
- Copilot Fusion tops Terminal Bench, beating Claude at less than half the cost — marlene_zw · 2026-11-18
- Microsoft announces MAI Code 2, plus Code 2 Flash and Cyber 2 Flash at MAI Ignite — testingcatalog · 2026-11-18
- Microsoft-Decision-1.1 hits GA with 2x speed at lower cost — marlene_zw · 2026-11-18
- Microsoft announces MAI Code 2 for GitHub Copilot — testingcatalog · 2026-11-18
- Microsoft Ignite: GPT-6.2 Astra and Sol land in Azure AI Foundry — testingcatalog · 2026-11-18
- Microsoft Decision 1 model officially announced — testingcatalog · 2026-11-18