Workspace-Bench 1.0: Benchmarking AI Agents on Workspace Tasks with Large-Scale File Dependencies

TL;DR

Workspace-Bench 1.0 introduces a large-scale, realistic benchmark for evaluating AI agents on workspace tasks involving complex file dependencies, highlighting current limitations in agent performance.

Contribution

The paper presents a new comprehensive benchmark with realistic workspaces, extensive tasks, and evaluation of multiple models, addressing the gap in workspace-level AI evaluation.

Findings

Current agents achieve only about 60% accuracy, below human performance of 80.7%.

Average agent performance across models is only 43.3%.

Workspace-Bench-Lite reduces evaluation costs by 70% while maintaining distribution.

Abstract

Workspace learning requires AI agents to identify, reason over, exploit, and update explicit and implicit dependencies among heterogeneous files in a worker's workspace, enabling them to complete both routine and advanced tasks effectively. Despite its importance, existing relevant benchmarks largely evaluate agents on pre-specified or synthesized files with limited real-world dependencies, leaving workspace-level evaluation underexplored. To this end, we introduce Workspace-Bench, a benchmark for evaluating AI agents on Workspace Learning involving Large-Scale File Dependencies. We construct realistic workspaces with 5 worker profiles, 74 file types, 20,476 files (up to 20GB) and curate 388 tasks, each with its own file dependency graph, evaluated across 7,399 total rubrics that require cross-file retrieval, contextual reasoning, and adaptive decision-making. We further provide Workspace-Bench-Lite, a 100-task subset that preserves the benchmark distribution while reducing evaluation costs by about 70%. We evaluate 4 popular agent harnesses and 7 foundation models. Experimental results show that current agents remain far from reliable workspace learning, where the best reaches only about 60%, substantially below the human result of 80.7%, and the average performance across agents is only 43.3%.