GPT-6.2 Sol100%
GPT-6.2 system cardattempts per vulnerability: 5 · capability flags: 16
OpenAI ↗Turns known V8 N-day vulnerabilities into exploit primitives against a hardened production-like engine.
| Model / creator | Version | Reported score |
|---|---|---|
| GPT-6.2 system card | 100% | |
| GPT-6.2 system card | 100% |
attempts per vulnerability: 5 · capability flags: 16
OpenAI ↗attempts per vulnerability: 5 · capability flags: 16
OpenAI ↗Tests multi-stage cybersecurity scenarios, helping authorized security teams assess an agent's ability to plan and execute complex assessments.
CVE RediscoveryTests finding previously documented vulnerabilities from vulnerable code snapshots, helping authorized security teams assess AI support for code audits.
Vision2WebTests building functional websites from visual prototypes and requirements, helping you compare models for turning a client's design into working code.