Models
Model comparisons
Side-by-side looks at the AI models that power agents — held to the same rules as everything else here: every claim links to a source, opinion is labeled, and corrections stay visible on the page.
Coming soon
The first comparison is in the works
Nothing is published here yet — comparisons only go live once every number in them can be traced to a source we can link. New comparisons will be listed on this page as they're ready.
In the meantime, the evaluation guide explains how model benchmarks work and what a score does (and doesn't) tell you — useful background for reading any comparison, including ours.