QuArch: A Benchmark for Evaluating LLM Reasoning in Computer Architecture
Abstract
The field of computer architecture, which bridges high-level software abstractions and low-level hardware implementations, remains absent from current large language model (LLM) evaluations. To this end, we present QuArch (pronounced 'quark'), the first benchmark designed to facilitate the development and evaluation of LLM knowledge and reasoning capabilities specifically in computer architecture. QuArch provides a comprehensive collection of 2,671 expert-validated question-answer (QA) pairs covering various aspects of computer architecture, including processor design, memory systems, and interconnection networks. Our evaluation reveals that while frontier models possess domain-specific knowledge, they struggle with skills that require higher-order thinking in computer architecture. Frontier model accuracies vary widely (from 34% to 72%) on these advanced questions, highlighting persistent gaps in architectural reasoning across analysis, design, and implementation QAs. By holistically assessing fundamental skills, QuArch provides a foundation for building and measuring LLM capabilities that can accelerate innovation in computing systems. With over 140 contributors from 40 institutions, this benchmark represents a community effort to set the standard for architectural reasoning in LLM evaluation.
Cite
@article{arxiv.2510.22087,
title = {QuArch: A Benchmark for Evaluating LLM Reasoning in Computer Architecture},
author = {Shvetank Prakash and Andrew Cheng and Arya Tschand and Mark Mazumder and Varun Gohil and Jeffrey Ma and Jason Yik and Zishen Wan and Jessica Quaye and Elisavet Lydia Alvanaki and Avinash Kumar and Chandrashis Mazumdar and Tuhin Khare and Alexander Ingare and Ikechukwu Uchendu and Radhika Ghosal and Abhishek Tyagi and Chenyu Wang and Andrea Mattia Garavagno and Sarah Gu and Alice Guo and Grace Hur and Luca Carloni and Tushar Krishna and Ankita Nayak and Amir Yazdanbakhsh and Vijay Janapa Reddi},
journal= {arXiv preprint arXiv:2510.22087},
year = {2025}
}