xbench: Tracking Agents Productivity Scaling with Profession-Aligned Real-World Evaluations

TL;DR

xbench is a novel evaluation suite that measures AI agents' productivity in real-world professional domains, aligning benchmarks with economic value and enabling tracking of capabilities over time.

Contribution

It introduces profession-aligned, real-world evaluation benchmarks for AI agents, bridging the gap between technical skills and economic productivity.

Findings

Initial benchmarks for Recruitment and Marketing domains.

Evaluation results for leading AI agents establish baseline performance.

xbench metrics correlate with productivity and predict Technology-Market Fit.

Abstract

We introduce xbench, a dynamic, profession-aligned evaluation suite designed to bridge the gap between AI agent capabilities and real-world productivity. While existing benchmarks often focus on isolated technical skills, they may not accurately reflect the economic value agents deliver in professional settings. To address this, xbench targets commercially significant domains with evaluation tasks defined by industry professionals. Our framework creates metrics that strongly correlate with productivity value, enables prediction of Technology-Market Fit (TMF), and facilitates tracking of product capabilities over time. As our initial implementations, we present two benchmarks: Recruitment and Marketing. For Recruitment, we collect 50 tasks from real-world headhunting business scenarios to evaluate agents' abilities in company mapping, information retrieval, and talent sourcing. For Marketing, we assess agents' ability to match influencers with advertiser needs, evaluating their performance across 50 advertiser requirements using a curated pool of 836 candidate influencers. We present initial evaluation results for leading contemporary agents, establishing a baseline for these professional domains. Our continuously updated evalsets and evaluations are available at https://xbench.org.