Model

BLOOM

Vendor N/A

Knowledge & Reasoning

Knowledge & Reasoning
BenchmarkScoreMetricHarnessSourceConfidenceLink
ARC-Challenge32.9accuracyN/AIndependentmediumarxiv.org
ARC-Easy40.7accuracyN/AIndependentmediumarxiv.org
BoolQ70.4scoreN/AIndependentmediumcrfm.stanford.edu
HellaSwag74.4overall_accuracyN/AIndependentmediumcrfm.stanford.edu
medqa23.3accuracyN/AIndependentmediumarxiv.org
PubMedQA73.6accuracyN/AIndependentmediumarxiv.org

Math

Math
BenchmarkScoreMetricHarnessSourceConfidenceLink
GSM8K9.5exact_matchN/AIndependentmediumcrfm.stanford.edu

Reproduced by the community

Loading community runs…