Google unveils Gemini 4 Carbon for finding and fixing flaws
Original title: Google Says Gemini 4 Carbon Leads in Finding, Exploiting and Fixing Vulnerabilities
Limited Fairwind access comes first; Google cites 88.2% on SEC-bench Pro and prices of $2 and $10 per million tokens.
Google announced Gemini 4 Carbon, a frontier model it says outperforms rivals at detecting, exploiting and patching vulnerabilities. It is not broadly available: access starts with selected cyber defenders, then approved partners in the Fairwind Program, which Google says includes more than 650 organizations. Google reported Carbon ahead of GPT-6 Astra and Claude models on its comparisons, with no guardrails, and said it will gather feedback before a wider launch. Introductory pricing matches Gemini 4 Argon at $2 per million input tokens and $10 per million output tokens.
Unconfirmed: Benchmark leads and the Chrome team’s patch comparison are Google’s own figures, not independently verified.
- Google says Carbon scored 54.8% on ExploitGym and 71.0% on CWE-bench.
- Gray Swan saw no successful indirect prompt-injection attacks against Carbon.
- After the intro period, prices rise to $4 and $20 per million tokens.
- Carbon follows Gemini 4 Argon, launched two weeks earlier.
- Organizations
- Wiz
- Countries
- United States
From the original article
2 min read at securityweek.com
Google’s new frontier model scored 88. 2% on SEC-bench Pro and is rolling out first to vetted cyber defenders via its Fairwind Program. The post Google Says Gemini 4 Carbon Leads in Finding, Exploiting and Fixing Vulnerabilities appeared first on SecurityWeek.
By Eduard Kovacs · securityweek.com
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.