jevIMAGE
Jev benchmarks against leading LLM judges on ScopeJudge
Dreadnode found that Jev catches agent scope violations, responding in about 130 milliseconds on average at a few cents per thousand checks.
dreadnode
@dreadnodeArchived 2026.09.22Originally posted 2026.09.22
dreadnode @dreadnodeOriginally posted 2026.09.22
Does Jev live up to the hype? Based on the results of running it against our ScopeJudge benchmark, it does. @typesafeai's Jev was competitive with leading LLM judges, catching agent scope violations at pennies per thousand checks, with 130 millisecond responses on average. [1/4]
x.com/dreadnode/status/2102131162386710885Archived 2026.09.22 10:12