Model

PaLM 2

Google

Knowledge & Reasoning

Benchmark	Score	Metric	Harness	Source	Confidence	Link
MMLU	78.6	exact_match	N/A	Independent	medium	crfm.stanford.edu

Benchmark	Score	Metric	Harness	Source	Confidence	Link
GSM8K	80.7	accuracy	N/A	Independent	medium	arxiv.org
MATH	34.3	accuracy	N/A	Independent	medium	arxiv.org

Benchmark	Score	Metric	Harness	Source	Confidence	Link
MGSM	72.2	average_(%)	N/A	Independent	medium	arxiv.org
XCOPA	94.4	accuracy	N/A	Independent	medium	arxiv.org