Model

J1-Jumbo v1

Vendor N/A

Knowledge & Reasoning

Knowledge & Reasoning
BenchmarkScoreMetricHarnessSourceConfidenceLink
BoolQ77.6scoreN/AIndependentmediumcrfm.stanford.edu
HellaSwag76.5overall_accuracyN/AIndependentmediumcrfm.stanford.edu

Math

Math
BenchmarkScoreMetricHarnessSourceConfidenceLink
GSM8K5.4exact_matchN/AIndependentmediumcrfm.stanford.edu

Reproduced by the community

Loading community runs…