Google 发布 Gemini 4 Carbon，安全基准超 GPT-6 Astra 与 Claude Opus 5.5
Google 两周内连发两代旗舰，Carbon 的漏洞发现和修补成绩直接压过 OpenAI 和 Anthropic，长上下文 96.2% 也很猛，定价还先打折，搞安全的建议盯一下。
Google 发布新旗舰模型 Gemini 4 Carbon，距上一代 Gemini 4 Argon 仅两周。在 SEC-bench Pro 漏洞发现上得分 88.2%，CWE-bench 修补得分 71.0%，ExploitGym 利用得分 54.8%。GraphWalks 长上下文基准上达到 96.2%，高于 Argon 的 84.2%、GPT-6 Astra 的 71.8% 和 Claude Opus 5.5 的 66.8%。定价为每百万 token 输入 2 美元、输出 10 美元，之后将涨至 4 美元和 20 美元。因具备漏洞利用能力，首发仅面向部分网络防御者与 Fairwind Program 合作伙伴。
Gemini 4 Carbon, Google's new frontier model, beats GPT-6 Astra and Claude Opus 5.5 on Artificial Analysis.
> Arriving 2 weeks after Gemini 4 Argon, Carbon gains most in security: 88.2% on SEC-bench Pro for finding vulnerabilities and 71.0% on CWE-bench for patching. 54.8% on ExploitGym for exploiting them and 71.0% on CWE-bench for patching them.
> Because of that exploit skill, access starts this week with a subset of cyber defenders, then trusted partners through the Fairwind Program, before developers, businesses and consumers get it.
> Google says Carbon can work on engineering problems for days, root-causing subtle bugs across large codebases instead of stalling at dead ends.
> Cost: introductory $2/$10 per 1M input/output tokens, later doubling to $4 and $20.
> On long-context benchmark: on GraphWalks (256K to 1M tokens, BFS F1), Carbon scores 96.2%, against Argon’s 84.2%, GPT-6 Astra’s 71.8% and Claude Opus 5.5’s 66.8%.
