← Anthropic · Claude Individual model view Anthropic · Claude

Claude Haiku 4.5

Appears in 6 benchmarksMean lean (confidence − pass rate): +0.004/6 benchmarks lean overconfident (prospective probe)

Positioning spread

0.000.250.500.751.00SQuAD (factual recall)MMLU-Pro (knowledge)LegalBench (legal reasoning)MathBench (competition math)OmniMath (advanced math)SciCode (scientific code)performanceconfidence (red gap = overconfident)
Performance vs. confidence for Claude Haiku 4.5, per benchmark (prospective probe).
BenchmarkTask accConfidenceF₁Leans
SQuAD (factual recall)0.480.400.72-0.08 cautious
MMLU-Pro (knowledge)0.650.670.74+0.03 calibrated
LegalBench (legal reasoning)0.860.640.74-0.23 cautious
MathBench (competition math)0.951.000.97+0.05 calibrated
OmniMath (advanced math)0.580.740.75+0.16 overconfident
SciCode (scientific code)0.560.650.73+0.09 overconfident

In the full cloud

-3-3-2-2-1-100112233Performance z-score within benchmark/probe →Confidence z-score →
Claude Haiku 4.5 conditions all other model/condition points equal relative confidence and pass rate

Pairwise footprint

Match accuracy controls for the performance base-rate gap
Claude Haiku 4.5 pairs
18/ 171
Claude Haiku 4.5 mean tau
+0.058
All-pairs mean
+0.037
Claude Haiku 4.5 p<0.05
9(50%)
-1.0-0.50.00.51.0Pair signal: do confidence gaps rank performance gaps? (Kendall tau-b)
all model pairs (observed) base-rate-matched null calibration-preserving null Claude Haiku 4.5 pair (filled = p<0.05) Claude Haiku 4.5 mean all-pairs mean
vs Claude 3 Haiku →vs Claude 3.5 Sonnet →vs Claude Sonnet 4.5 → Compare with anything →