§ feed · storyline

WANDR Benchmark: Evaluating Research Agents That Must Search Wide and Deep

Perplexity releases WANDR benchmark for evaluating research agents' wide and deep search performance.

Jul 14 · 02:00:00 · primary fetch1 sourceupdated Jul 14 · 02:00:00

Perplexity released findings from the WANDR benchmark, which evaluates the performance of various research agents in wide and deep searching.

§ sources1 publication · timeline below