SEC-bench Pro
Named by 7 kept items in this workspace.
4 items name it under tools and 3 under classification.key entities.
Search every workspace for SEC-bench Pro. It matches the name as written, so it can find other things that share it.
The items that name it 7
benchmark testing a model's ability to uncover flaws in real-world software
Evaluates a model's ability to discover vulnerabilities in large JavaScript engines, including V8 and SpiderMonkey. GPT-6 Luna (October 29) scored 51.75%, compared with 85.4% for GPT-6 Astra
evaluates models' ability to discover vulnerabilities in large JavaScript engines, including V8 and SpiderMonkey
evaluates models' ability to discover vulnerabilities in large JavaScript engines, including V8 and SpiderMonkey
All technologies in this workspace.