JevBench 1.4 leaderboard

JevBench 1.4 is a reasoning benchmark published by JevBench v1.4 method. Decision-system evaluation combining chance-corrected intelligence, probability calibration, latency, and cost across 842 decisions. Its evaluation set is described by the source as 534 frozen v1.2 decisions plus 308 sealed v1.4 decisions. The source rates its difficulty as typed decisions with sealed generalization tests.

Each model's result is published by JevBench v1.4 method as equal-weight harmonic composite with gates. ModelScale prints that value to 2 decimal places, exactly as published, and never converts or rescales it. This page ranks the 184 models that carry a published value, highest value first; a model the source never scored is left out rather than ranked at zero. The source does not state whether a higher value is the better result, so neither does this page.