ITBench: Evaluating AI Agents across Diverse Real-World IT Automation Tasks
Saurabh Jha, Rohan Arora, Yuji Watanabe, Takumi Yanagawa, Yinfang Chen, Jackson Clark, Bhavya Bhavya, Mudit Verma, Harshit Kumar, Hirokuni Kitahara, Noah Zheutlin, Saki Takano, Divya Pathak, Felix George, Xinbo Wu, Bekir O. Turkkan, Gerard Vanloo, Michael Nidd, Ting Dai, Oishik Chatterjee, Pranjal Gupta, Suranjana Samanta, Pooja Aggarwal, Rong Lee, Pavankumar Murali, Jae-wook Ahn, Debanjana Kar, Ameet Rahane, Carlos Fonseca, Amit Paradkar, Yu Deng, Pratibha Moogi, Prateeti Mohapatra, Naoki Abe, Chandrasekhar Narayanaswami, Tianyin Xu, Lav R. Varshney, Ruchi Mahindru, Anca Sailer, Laura Shwartz, Daby Sow, Nicholas C. M. Fuller, Ruchir Puri
TL;DR
ITBench presents the first comprehensive, open-source benchmark framework for evaluating AI agents across diverse real-world IT automation tasks, spanning SRE, CISO, and FinOps. It couples a 94-scenario benchmark with a push-button deployment workflow, multi-modal observability tools, and a leaderboard that supports partial scoring for constructive feedback. Empirical results show that state-of-the-art models achieve limited success (e.g., GPT-4o attaining the best but still modest pass@1 scores, with performance dropping as scenario complexity increases and under limited observability). The framework emphasizes reproducibility, extensibility, and safety, and aims to catalyze progress toward correct, safe, and fast AI-driven IT automation in production environments.
Abstract
Realizing the vision of using AI agents to automate critical IT tasks depends on the ability to measure and understand effectiveness of proposed solutions. We introduce ITBench, a framework that offers a systematic methodology for benchmarking AI agents to address real-world IT automation tasks. Our initial release targets three key areas: Site Reliability Engineering (SRE), Compliance and Security Operations (CISO), and Financial Operations (FinOps). The design enables AI researchers to understand the challenges and opportunities of AI agents for IT automation with push-button workflows and interpretable metrics. ITBench includes an initial set of 94 real-world scenarios, which can be easily extended by community contributions. Our results show that agents powered by state-of-the-art models resolve only 13.8% of SRE scenarios, 25.2% of CISO scenarios, and 0% of FinOps scenarios. We expect ITBench to be a key enabler of AI-driven IT automation that is correct, safe, and fast.