Model

Anthropic-LM v4-s3

Vendor N/A

Knowledge & Reasoning

Knowledge & Reasoning
BenchmarkScoreMetricHarnessSourceConfidenceLink
BoolQ81.5scoreN/AIndependentmediumcrfm.stanford.edu
HellaSwag80.7overall_accuracyN/AIndependentmediumcrfm.stanford.edu

Math

Math
BenchmarkScoreMetricHarnessSourceConfidenceLink
GSM8K17.1exact_matchN/AIndependentmediumcrfm.stanford.edu

Reproduced by the community

Loading community runs…