Announcement
Google Releases Gemini 4 Carbon Frontier Model, Reportedly #1 on Artificial Analysis Index with Cybersecurity Focus
A post says Carbon scored 88.2% on SEC-bench Pro and is rolling out first to cybersecurity specialists.
TLDR
A post calls Gemini 4 Carbon Google's second frontier model in about a month, says it leads the Artificial Analysis Intelligence Index v5 and cites an 88.2% SEC-bench Pro score. It says the model can work on complex coding tasks for days and will reach cybersecurity defenders first, with wider access planned. Another post relays Google's internal claim that Carbon writes secure code twice as well as Argon.
Combined views
3.3M
68 Sources, first seen ago
