Skip to main content

Haiku + Gemini + GPT-4o mini, persona

Ensemble of Haiku 4.5, Gemini 2.5 Flash-Lite, GPT-4o mini (all persona)

OpinionsQA · Ensemble · default settings · 684 questions

ensemble--3-model-blend--tdefault--tplcurrent--1bef3e62

Compare with another OpinionsQA config → Haiku + Gemini + GPT-4o mini, persona profile →

Metadata

Ensemble · 3 replicates · 684 questions
Method
Ensemble
Base provider
—
Model
3-model-blend
Dataset
OpinionsQA
Temperature
default
Reasoning effort
default
Template
current
Samples / question
—
Questions
684
First replicate
2026-04-12 02:07 UTC
Last replicate
2026-10-13 20:13 UTC

Aggregated metrics

Mean across replicates, ± sample standard deviation (shown only with 2+ replicates). SynthBench Index: 0 = random answers, 100 = real people. SPS is the raw 0 to 1 score it is computed from. Index = 100 × (SPS − SPS of random answers) ÷ (human ceiling − SPS of random answers), on this config's dataset. The leaderboard averages it over the core datasets.

Index
0 = random answers, 100 = real people
47
± 1.3
SPS
SPS, 0 to 1
0.875
± 0.003
Distribution gap
mean JSD, lower is better
0.176
± 0.016
Rank agreement
Kendall τ, −1 to 1
0.659
± 0.026

Variance summary

Replicates
3
Std dev
1.3 Index points (SPS 0.003)
Range
Index 46 to 48 (SPS 0.871–0.877)
Coefficient of variation (SPS)
0.0036

Per-topic breakdown

Index per topic (mean across replicates; 0 = random answers, 100 = real people). All seven topics are shown; topics this config's questions don't cover are marked "no questions".

Show data table

Loading data…

Replicates (3)

Run Timestamp Index SPS JSD Kendall τ N
opinionsqa_ensemble_3blend_20260412_020745 2026-04-12 02:07 UTC 48 0.877 0.167 0.674 684
opinionsqa_ensemble_3blend_20260412_021342 2026-04-12 02:13 UTC 48 0.877 0.167 0.674 684
opinionsqa_ensemble_3blend_20261013_201306 2026-10-13 20:13 UTC 46 0.871 0.195 0.629 100