OpenVibeEval · Web works
OpenVibeEval · Compare web work, accessibility and blind preference to observe design performance across different tools.
What it measures, and how
Models and harnesses are stored separately; accessibility scores are not aesthetic scores, and independent blind-selection results and identical run conditions must be verified before adoption.
How this evidence is used
Under observation: raw preference data from blind tests under the same conditions and in sufficient volume, plus a publication time, are needed.
Limits and data attribution
Models mix different run tools; the main table is accessibility and task outcomes, not a unified aesthetic vote.
Data licence: Stable score export and public aggregation permission are unconfirmed.
Scores published by OpenVibeEval. Raw scores and the News consensus score use different scales and cannot be added directly.