Describe a page. See how each model builds it.

Your prompt goes to every contender at once. Compare the results side by side, with cost, latency, and token counts for each.