Evidence-Based
Contributes to:Informativeness
Claude Fable 5.1 (Max, Default Fallback) ranks highest on evidence-based, while GPT-3.5 Turbo ranks lowest across the 42 models evaluated. The gap between them is 1895 Elo.
Elo scores span −105 to 1790, with a typical per-model 95% confidence interval of ±31 Elo, roughly ±4 points on the 1 to 100 scale.
Evidence-Based Leaderboard
Each model’s Elo on this trait, percentile ranked from 1 to 100 across all models evaluated in the benchmark. Whiskers show each model’s 95% confidence interval on the same scale.
AA-PersonalityBench: Evidence-Based
Does the response ground claims in evidence, data, or authoritative sources? · Trait Elo percentile ranked from 1 to 100 across all models evaluated in the benchmark
Other Traits
Lexically VariedEloquentLiterateNatural SoundingMetaphoricalStructuredScannableBottom-Line SummaryAnswer FirstInformation DenseResource-SpecificActionable GuidancePragmaticTruthfulAdmits UncertaintyDisclaims AssumptionsFlags NuancesSycophanticFriendlyHumorousSincereEmotionally IntelligentDirectConfidentOpinionatedConstructive PushbackCreativeOpen-EndedSlop