Agents' Last Exam

Researchers introduced Agents' Last Exam (ALE), a new benchmark designed to evaluate AI agents on long-horizon, economically valuable, real-world tasks with verifiable outcomes. Developed with over 250 industry experts, the benchmark covers 1,000+ tasks across 55 subfields in 13 industry clusters, with current results showing an average full pass rate of just 2.6% on the hardest tier. The benchmark aims to close the gap between AI benchmark success and GDP-relevant economic impact by serving as a continuously updated instrument for measuring sustained performance on real professional workflows.

arXiv:2606.05405v1 Announce Type: new Abstract: Recent AI systems have achieved strong results on a wide range of benchmarks, yet these gains have not translated into economically meaningful deployment across many professional domains. We argue that this gap is largely an evaluation problem: widely used benchmarks lack sustained performance measurement on real and economically valuable workflows. This paper introduces Agents' Last Exam ALE , a benchmark designed to evaluate AI agents on long-horizon, economically valuable, real-world tasks with verifiable outcomes. Developed in collaboration with 250+ industry experts, ALE covers non-physical industries defined with reference to O NET / SOC 2018 the U.S. federal occupational taxonomy . It is organized around a task taxonomy with 55 subfields grouped into 13 industry clusters covering 1K+ tasks. Current results show that the hardest tier remains far from saturated: across mainstream harness and backbone configurations, the average full pass rate is 2.6%. ALE is designed as a living benchmark: its task pool grows continuously as new workflows and industries are onboarded. More broadly, ALE is intended not merely as another leaderboard, but as an instrument for closing the gap between benchmark success and GDP-relevant impact.