Artificial Analysis: for Claude, a bigger model beats maxed-out effort on intelligence per dollar

Artificial Analysis says that with the current Claude lineup, moving up a model tier gets more intelligence per dollar than raising reasoning effort to the maximum. In its chart of performance against cost, it reports that at nearly every cost per task the largest model, Claude Opus 5.5, sits on the Pareto frontier for Anthropic's models.

This is Artificial Analysis's own reading of its measurements; the post gives no figures beyond the chart.