Model

llama 2

Meta · open weights

Knowledge & Reasoning

Benchmark	Score	Metric	Harness	Source	Confidence	Link
medqa	59.2	accuracy	N/A	Independent	medium	arxiv.org
MMLU	73.5	average_(%)	N/A	Independent	medium	arxiv.org
MultiTQ	18.5	hits@1	N/A	Independent	medium	arxiv.org

Benchmark	Score	Metric	Harness	Source	Confidence	Link
MAWPS	82.4	accuracy_(%)	N/A	Independent	medium	arxiv.org
SVAMP	69.2	execution_accuracy	N/A	Independent	medium	arxiv.org

Benchmark	Score	Metric	Harness	Source	Confidence	Link
rt-inod-jailbreaking	86.0	best-of	N/A	Independent	medium	arxiv.org