Vision

Rows. Each model once, at its highest reasoning level with a vision score; every level is in the table below the charts. A model that cannot take images scores 0 in the index, labelled, and is listed in that table, not drawn here.

What this metric means. The plain average of the two source scores: the share of questions answered correctly from each source, with no tools and no code.

How to read this chart. Each point is one model at its highest reasoning level. Further left means a lower average cost per question (log scale); higher means a higher vision score. Both axes fit the points shown, so the score axis does not start at zero. The shaded most attractive quadrant is the top-left quarter of the plot, split at the middle of each axis: a reading aid, not a fixed bar. The dotted Pareto line joins the points that no cheaper one beats on score. Only exact costs are plotted; the table lists every entry.

Cost per question. What one question costs at the provider's published API prices: one model call with the question's images, averaged over the questions in the score. A model with any question whose cost was not recorded is not plotted.