Model

Qwen3 1.7B

Alibaba · open weights

qwenlm.github.io

Knowledge & Reasoning

Knowledge & Reasoning
BenchmarkScoreMetricHarnessSourceConfidenceLink
GPQA Diamond38.0accuracyN/AIndependenthighepoch.ai
OTel52.8scoreN/AIndependentmediumarxiv.org
PhysicsMate5.5accuracylora adaptationIndependentmediumarxiv.org

Safety & Alignment

Safety & Alignment
BenchmarkScoreMetricHarnessSourceConfidenceLink
Guise78.4attack_success_rateN/AIndependentmediumarxiv.org
PFaithBench48.2overansN/AIndependentmediumarxiv.org

Multilingual

Multilingual
BenchmarkScoreMetricHarnessSourceConfidenceLink
sk-bench35.6scoreN/AIndependentmediumarxiv.org

Reproduced by the community

Loading community runs…