Model

Qwen3.5-0.8B

Alibaba · open weights

qwenlm.github.io

Math

Math
BenchmarkScoreMetricHarnessSourceConfidenceLink
GraphDecide70.6accuracyN/AIndependentmediumarxiv.org

Multilingual

Multilingual
BenchmarkScoreMetricHarnessSourceConfidenceLink
sk-bench33.2scoreN/AIndependentmediumarxiv.org

Reproduced by the community

Loading community runs…