| title | Φ-Bench: Can Large Language Models Engineer the Infrastructure That Powers Them? |
|---|---|
| authors | Leilei Ding, Shumin Wang, Yuting Huang, Fanqi Wan, Yinmin Zhang, Qi Han, Yiming Xu, Feiyuan Zhang, Xiaomeng Chu, Guoliang You, Wuyang Zhang, Daxin Jiang, Yanyong Zhang |
| date | 2026-09-09 |
| venue | arXiv (preprint, not peer reviewed) |
| id | arXiv:2609.10226 |
| affiliation | StepFun |
| upvotes | 16 (Hugging Face Daily Papers, 2026-11-01) |
| links | abs pdf hf code |
| abstract | Large language models (LLMs) have demonstrated remarkable capabilities in reasoning and code generation, raising the prospect that they could assist in developing and optimizing the very infrastructure that powers them. However, existing benchmarks mainly focus on isolated kernels, predefined operators, or pre-specified optimization targets, and therefore fail to evaluate the ability of LLMs to perform open-ended, long-horizon LLM infrastructure engineering. To address this gap, we present Φ-Bench, a benchmark for systematically evaluating LLMs on engineering the LLM infrastructure stack. Derived from optimization problems studied in frontier research and grounded in real-world code repositories, Φ-Bench provides broad coverage of the LLM infrastructure stack and spans tasks of varying complexity, ranging from localized kernel-level function completion to long-horizon implementation and end-to-end system optimization. Extensive experiments on frontier LLMs reveal their current capabilities and limitations in engineering complex LLM infrastructure, offering insights into the challenges that remain on the path toward autonomous optimization of future AI infrastructure. |
cite
@misc{ding2026bench,
title={Φ-Bench: Can Large Language Models Engineer the Infrastructure That Powers Them?},
author={Leilei Ding and Shumin Wang and Yuting Huang and Fanqi Wan and Yinmin Zhang and Qi Han and Yiming Xu and Feiyuan Zhang and Xiaomeng Chu and Guoliang You and Wuyang Zhang and Daxin Jiang and Yanyong Zhang},
year={2026},
eprint={2609.10226},
archivePrefix={arXiv},
url={https://arxiv.org/abs/2609.10226}
}