BenchmarkSEC Bench Pro
SEC Bench Pro
A vulnerability-discovery benchmark oriented toward real code repositories, assessing whether a model can reliably locate known or injected defects within a given repository.
Core Facts
Timeline (last 90 days)
Oct 29
The SEC Bench Pro score is 88.2%.
Unverified50%
Oct 29
On the SEC Bench Pro benchmark, Carbon achieved 88.2%, surpassing GPT-6 Astra's 85.4% and Argon's 83.4%.
Unverified50%
Oct 29
Carbon scored 88.2% on SEC Bench Pro, setting a new record.
Unverified50%
Oct 28
Gemini 4 Carbon achieved 88.2% on the SEC-bench Pro benchmark.
Partially Verified65%
All Facts (4)
Partially Verified
Gemini 4 Carbon achieved 88.2% on the SEC-bench Pro benchmark.
65%UnverifiedThe SEC Bench Pro score is 88.2%.
50%UnverifiedOn the SEC Bench Pro benchmark, Carbon achieved 88.2%, surpassing GPT-6 Astra's 85.4% and Argon's 83.4%.
50%UnverifiedCarbon scored 88.2% on SEC Bench Pro, setting a new record.
50%Source Articles
Gemini 4 Carbon Released: Tops Third-Party Leaderboard, But Ordinary Developers Still Left OutOct 29Google Retakes Global #1 in Two Weeks: Gemini 4 Carbon ReviewOct 29Gemini 4 Carbon Released: The Flagship Model That Overtakes Opus 5.5 Across the BoardOct 29Gemini 4 Carbon Released: The Strongest Cybersecurity Model, But You Can't Use It YetOct 29Google Releases Gemini 4 Carbon: Security Agent Back on TopOct 29