Checked fact 309829 Oct 2026Models and releases

Google reports Carbon "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."

By ai notis. An AI newsroom with a human editor. Every fact links to its source. How we work

The exact words it rests on

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.

What the source said when we opened it, on 29 Oct 2026.

The source

Gemini 4 Carbon: our next frontier model
Google · 2026-10-28

Open the source archived copy

Checked

Checked by the notis newsroom on 29 Oct 2026, against the source above.

The number in it

88.2% · Gemini 4 Carbon, benchmark score (SEC-bench Pro)
54.8% · Gemini 4 Carbon, benchmark score (ExploitGym)
71% · Gemini 4 Carbon, benchmark score (CWE-bench)

In the story

Gemini 4 Carbon opens to some cyber defenders first, with no developer date 29 Oct 2026

Cite this fact

Short permanent address

In a sentence

“Google reports Carbon "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.".” ai notis, checked 29 Oct 2026. Source: Gemini 4 Carbon: our next frontier model (Google). https://ainotis.com/c/3098

As a reference (APA style)

ai notis. (2026, October 29). Google reports Carbon "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." [Checked claim 3098]. ai notis. https://ainotis.com/c/3098

Anyone may quote this address. It does not change; if we correct the story, this page says so.