Benchmark

CVE-Bench

What is CVE-Bench?

Real CVE exploitation in sandboxes. Vendors report it as a safety ceiling rather than a capability to maximize.

Released: 2025-03-21
Evaluates: security
Openness: unknown
Importer: Model reports
Reported scores: 0

Source provenance

Original evidence https://github.com/uiuc-kang-lab/cve-bench

CVE-Bench code

Code https://github.com/uiuc-kang-lab/cve-bench

No reported scores are on record for this benchmark yet.

Which sources cite CVE-Bench?

GPT-5.6 (Sol, Terra, Luna) OpenAI, 2026-07-09

Related security benchmarks