Perplexity AI has open-sourced WANDR, a benchmark designed to evaluate how well AI agents handle complex, multi-layered research tasks, giving developers a standardized way to measure investigative work. WANDR, which stands for Wide and Nuanced Deep Research, is built for “wide research” tasks requiring broad searching and deep investigation, aligned with Perplexity Computer’s capabilities. The benchmark mirrors demanding professional research workloads, evolving from earlier frameworks like WideSearch. It arrives alongside Perplexity’s “Search as Code” framework (SaC), which treats research queries as programmable workflows. In preliminary evaluations on June 1, 2026, Perplexity’s SaC implementation scored 0.386 on the WANDR benchmark, outperforming the next-best score of 0.152 by a factor of roughly 2.5x. WANDR is directly tied to Perplexity Computer, the company’s AI agent product that orchestrates multiple models. As of July 11, 2026, xAI’s Grok 4.5 achieved top scores on WANDR when used with Perplexity Computer.