Model

DeepSeek-V3.1

DeepSeek · open weights

deepseek.com

Knowledge & Reasoning

Knowledge & Reasoning
BenchmarkScoreMetricHarnessSourceConfidenceLink
PerSo67.5accuracyN/AIndependentmediumarxiv.org

Coding

Coding
BenchmarkScoreMetricHarnessSourceConfidenceLink
ARCS23.5ex_e2eN/AIndependentmediumarxiv.org

Safety & Alignment

Safety & Alignment
BenchmarkScoreMetricHarnessSourceConfidenceLink
BrowseSafe++79.1f1N/AIndependentmediumarxiv.org

Reproduced by the community

Loading community runs…