Google 发布 Gemini 4 Carbon，称其超越两周前的 Argon，位列 Artificial Analysis V5 第一。模型将率先向网络防御者推送，在 SEC-bench Pro 漏洞发现（88.2%）、ExploitGym 利用（54.8%）、CWE-bench 修补（71.0%）等网络安全基准上领先，并在编码和 Agent 任务上有显著提升。
原文给出 Gemini 4 Carbon 在多项安全与编码基准的具体得分，并附与 GPT-6 Astra、Claude 等模型的对比表，可供评估其实力。
BREAKING 🔥: Google announced Gemini 4 Carbon, a new SOTA model that takes the top spot on Artificial Analysis V5 ranking!
> Gemini 4 Carbon is rolling out to cyber defenders first.
> It shows significant gains across coding and agentic tasks as well as across key capabilities critical for cybersecurity defense, landing a firm lead in SEC-bench Pro (88.2%) for finding vulnerabilities, ExploitGym (54.8%) for exploitation, and CWE-bench (71.0%) for patching
It needs testing! 👀
Introducing Gemini 4 Carbon, it's a big jump over Argon which we released just 2 weeks ago. Gemini 4 Carbon is SOTA on cyber, long horizon tasks, ranks #1 on Artificial Analysis V5, significantly improves on usability in domains like coding, and is more aligned.在 X 查看被引用的帖子
来源：🚨 AI News | TestingCatalog · x.com