TL;DR: OpenAI has launched GPT-6 Astra, a new flagship model boasting enhanced intelligence, alignment, and critical cybersecurity capabilities.
Summary: OpenAI has released GPT-6 Astra, positioning it as their most intelligent and highly aligned model to date. Astra achieves Critical-level cybersecurity capabilities, scoring 100% on ExploitBench and identifying two zero-day vulnerabilities during evaluation. It sets new benchmarks across various domains, including FrontierMath Tier 4, ARC-AGI-3, Terminal-Bench Science, Agents' Last Exam, BenchCAD, and GPQA Diamond.
Why it matters: This release signifies a major leap in AI model capabilities, particularly in security and complex reasoning. AI developers should explore Astra's advanced performance in computer operations, coding, and scientific benchmarks for building more robust and secure applications.
Source: rss