Benchmarks

ExploitBench

Agentsunit % · normalized over [0, 100]

ExploitBench — turns known, disclosed vulnerabilities into working exploits end to end, the step above vulnerability discovery on the exploitation chain. Public-set scores are contamination-sensitive: labs increasingly report private internal ports alongside them.

#ModelSourceScoreNormalized

Score distribution

7 tracked results across the normalization window

0100

Score vs. parameters

Open-weights models, log-x params