Model

Qwen3 14B

Alibaba · open weights

qwenlm.github.io

Knowledge & Reasoning

Knowledge & Reasoning
BenchmarkScoreMetricHarnessSourceConfidenceLink
CritPt0.0accuracyN/AIndependentlowepoch.ai
GPQA Diamond63.8accuracyN/AIndependenthighepoch.ai
MS-Exam-Gen83.3accuracyN/AIndependentmediumarxiv.org
OTel60.7scoreN/AIndependentmediumarxiv.org

Coding

Coding
BenchmarkScoreMetricHarnessSourceConfidenceLink
SciCode31.6scoreN/AIndependentmediumartificialanalysis.ai

Safety & Alignment

Safety & Alignment
BenchmarkScoreMetricHarnessSourceConfidenceLink
PFaithBench26.2overansN/AIndependentmediumarxiv.org

Abstract Reasoning

Abstract Reasoning
BenchmarkScoreMetricHarnessSourceConfidenceLink
DrillBench51.5accuracylikelihood + pmiIndependentmediumarxiv.org

Reproduced by the community

Loading community runs…