← Google · Gemini Individual model view Google · Gemini

gemini-3-pro-preview

Appears in 1 benchmarksMean lean (confidence − pass rate): +0.031/1 benchmarks lean overconfident (prospective probe)

Positioning spread: every benchmark, one model

0.000.250.500.751.00SQuAD (factual recall)performanceconfidence (red gap = overconfident)
Performance vs. confidence for gemini-3-pro-preview, per benchmark (prospective probe).
BenchmarkTask accConfidenceF₁Leans
SQuAD (factual recall)0.650.680.79+0.03 calibrated

In the full cloud

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

Pairwise signal: pairs involving gemini-3-pro-preview

Match accuracy controls for the performance base-rate gap
gemini-3-pro-preview pairs
18/ 171
gemini-3-pro-preview mean tau
+0.086
All-pairs mean
+0.037
gemini-3-pro-preview p<0.05
16(88.9%)
-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 gemini-3-pro-preview pair (filled = p<0.05) gemini-3-pro-preview mean all-pairs mean

The four metacognitive outcomes

No curated cases for this selection yet — outcome-matrix extraction currently covers a sample of MMLU-Pro trials.
vs gemini-2.0-flash-001 →vs gemini-2.5-flash →vs gemini-2.5-pro →vs gemini-3-flash-preview →vs gemini-3.1-pro-preview → Compare with anything →