CategoryReasoning
MeasureOfficial option A harmonic composite with published uncertainty
Tasks904 open decisions plus 720 sealed decisions per complete run
DifficultyChoice, Noul, and Score decisions with sealed generalization tests
Typed-decision evaluation across 1,624 decisions, combining chance-corrected intelligence, probability calibration, latency, and cost.
JevBench 1.5 ranking
214 models with a published JevBench 1.5 value, ordered by that value, highest first. Models the source has not scored on this benchmark are not listed — they are unmeasured, not last.
About JevBench 1.5
What the benchmark measures and how its value is published, from the source’s own record.
What is JevBench 1.5?
JevBench 1.5 is a reasoning benchmark published by JevBench v1.5 frozen method and disclosures. Typed-decision evaluation across 1,624 decisions, combining chance-corrected intelligence, probability calibration, latency, and cost. Its evaluation set is described by the source as 904 open decisions plus 720 sealed decisions per complete run. The source rates its difficulty as choice, Noul, and Score decisions with sealed generalization tests.
How is JevBench 1.5 scored?
Each model's result is published by JevBench v1.5 frozen method and disclosures as official option A harmonic composite with published uncertainty. ModelScale prints that value to 2 decimal places, exactly as published, and never converts or rescales it. This page ranks the 214 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.
ModelScale
