Atlas

Benchmarks

VibePrevisBench Screen Check

VibePrevisBench's judged half: a Gemini 4 judge checks whether each task's ground-truth facts read on screen in the rendered proxy video, averaged over the 242 tasks. The paper finds the judge favours Gemini models slightly and other judges score lower, so it is shown as a diagnostic.

Models
9
Top score
85.4
Leader
GPT-6 Astra
Updated
Nov 11, 2026
Resolution
—
Saturation
—

No signal or difficulty estimate. Held out of the fit by design.

Loading Atlas data…