What it measures
Large-scale blind pairwise preference arena segmented by real-world user profession, education level, and language.
Reading the measurement
A preference result depends on the opponents, judges and prompts in the comparison pool. Use it to study which outputs people or judges favor, then check correctness and reliability with a separate task-based evaluation. Ratings from different arenas do not share a universal scale.
Published evaluation
Bradley–Terry preference ratings with controls for loading time, length and Markdown. Table 1 is a dated 13-model snapshot. Source confidence intervals are retained; source rank groups overlap and should not be read as certain ordering. Extended-thinking variants remain explicitly named.
Where this fits in an evaluation suite
Format compliance, preference, truthfulness, and safety require separate measurements. Keep desirable writing style from masking factual errors.
Start with a capability hypothesis, then choose a second test with a different failure mode. Keep the checkpoint, reasoning effort, agent tools, sampling policy, and scoring procedure fixed when measuring a training intervention. Record failures and missing outputs in the denominator.
A reproducible comparison
Bradley–Terry preference ratings with controls for loading time, length and Markdown. Table 1 is a dated 13-model snapshot. Source confidence intervals are retained; source rank groups overlap and should not be read as certain ordering. Extended-thinking variants remain explicitly named.
Publish the exact run configuration, item coverage, score units, and available uncertainty. If a source supplies only an aggregate or a single run, preserve that limit. A new model name or a rounded score alone does not establish a capability improvement.
PRIMARY EVIDENCE & FURTHER READING
Reviewed 2026-10-27. Explanation and interpretation by RLTasks. Benchmark ownership remains with its original authors.