all the models — AI benchmark observatory
← Benchmarks

CathedralBench

CathedralBench is a third-party red-team cyber evaluation on highly difficult, multiple-exploit-chain tasks in egress-secure sandboxes. Accuracy is commonly reported on the hard subset of tasks in an unrestricted configuration.

id cathedralbench · max 1 · 1 models reported

#ModelScore

No scores for this benchmark yet.