Skip to main content

Claude Haiku 4.5, raw

OpinionsQA · Raw LLM · default settings · 200 questions

raw--claude-haiku-4-5--tdefault--tplcurrent--d14c83b2

Compare with another OpinionsQA config → Claude Haiku 4.5 profile →

Metadata

Raw LLM · 3 replicates · 200 questions
Method
Raw LLM
Base provider
anthropic
Model
claude-haiku-4-5
Dataset
OpinionsQA
Temperature
default
Reasoning effort
default
Template
current
Samples / question
50
Questions
200
First replicate
2026-04-10 23:25 UTC
Last replicate
2026-04-10 23:50 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
22
± 0.5
SPS
SPS, 0 to 1
0.815
± 0.001
Distribution gap
mean JSD, lower is better
0.309
± 0.002
Rank agreement
Kendall τ, −1 to 1
0.522
± 0.005

Variance summary

Replicates
3
Std dev
0.5 Index points (SPS 0.001)
Range
Index 21 to 22 (SPS 0.814–0.816)
Coefficient of variation (SPS)
0.0013

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_openrouter_anthropic_claude-haiku-4-5_20260410_232509 2026-04-10 23:25 UTC 22 0.816 0.307 0.522 200
opinionsqa_openrouter_anthropic_claude-haiku-4-5_20260410_233815 2026-04-10 23:38 UTC 21 0.814 0.311 0.518 200
opinionsqa_openrouter_anthropic_claude-haiku-4-5_20260410_235056 2026-04-10 23:50 UTC 22 0.815 0.310 0.527 200