Benchmarks and tools/benchmark

SEC-bench Pro

Benchmark of long-horizon proof-of-concept generation for curated disclosed vulnerabilities, cited in Google’s Gemini 4 Carbon capability report.

Records citing SEC-bench Pro

Oct 28, 2026

Google announces Gemini 4 Carbon cyber results and phased access for defenders

CapabilityEvaluation reportGoogle DeepMind

Google announces Gemini 4 Carbon with vendor-reported scores of 88.2% on SEC-bench Pro, 54.8% on ExploitGym and 71.0% on CWE-bench, evaluated without guardrails. It describes phased access for selected cyber defenders through Fairwind, with broader product availability planned after further safeguard work.