Model

T5

Vendor N/A

Knowledge & Reasoning

Knowledge & Reasoning
BenchmarkScoreMetricHarnessSourceConfidenceLink
BoolQ81.4scoreN/AIndependentmediumarxiv.org
SuperGLUE76.2scoreN/AIndependentmediumarxiv.org

Math

Math
BenchmarkScoreMetricHarnessSourceConfidenceLink
GSM8K2.3exact_matchN/AIndependentmediumcrfm.stanford.edu

Reproduced by the community

Loading community runs…