Model

Mistral Large 4.0

Mistral AI · open weights

mistral.ai

Knowledge & Reasoning

Knowledge & Reasoning
BenchmarkScoreMetricHarnessSourceConfidenceLink
CritPt10.6accuracyN/AIndependentlowepoch.ai
GPQA Diamond88.1accuracyN/AIndependenthighepoch.ai
SimpleQA Verified20.0accuracyN/AIndependenthighepoch.ai

Coding

Coding
BenchmarkScoreMetricHarnessSourceConfidenceLink
SciCode54.2scoreN/AIndependentmediumartificialanalysis.ai

Reproduced by the community

Loading community runs…