MatOS: An Agent Operating System for Materials Research
Abstract
Materials research relies on scattered and growing data, knowledge, and simulation tools, and a single study often spans multi-day workflows across literature analysis, computation, and experiment. Large language model agents can now retrieve materials data, operate simulation codes, and close design-synthesis-characterization loops with autonomous laboratories. However, extending these agents with new capabilities, keeping long-running tasks on track, and letting them learn from past tasks and user feedback remain difficult. Here we present MatOS, an agent operating system for materials research with three core components: (1) a layered scientific asset graph that organizes data, knowledge, tools, skills, roles, and workflows as reusable assets; (2) a stateful adaptive workflow that runs tasks as explicit execution graphs with scientific checks, so that they can be resumed and traced; and (3) evidence-driven asset evolution, which revises these assets from execution records and user feedback. On ResearchClawBench, MatOS improves on the previously best-performing agent by 11.15% with the same base model and extends the cost-score Pareto frontier. With a common base model, it outperforms Codex CLI and Claude Code, exceeding Codex CLI by 2.13% on MatTools and 4.17% on SciAgentGym. In eight research cases, MatOS narrowed more than 30 million candidate structures to 97 solid-electrolyte candidates, identified 89 halide variants with higher computed conductivity than their parent structures, linked the force errors of 11 pretrained interatomic potentials to local atomic environments, and supported two new methods that reduce errors by 3.1% and 13.6% relative to their baselines. These results show that organizing research capabilities as reusable, evolving assets enables agents to carry out long, multistage materials studies. Code is available at https://github.com/DingyangLyu/MatOS.
Related
- Operating regimes and materials limits of nonlinear magnon dynamics in Co-doped YIG
- An LLM agent for end-to-end computational materials discovery
- Organizational Learning in Materials Discovery Campaigns: Which Databases, Calculations, and Agent Teams Pay Off
- FUSION: a skill-based research agent for publicly obtainable nuclear-physics codes
- Building a research-software catalog with a coding agent: from hackathon prototype to public deployment
BibTeX
@misc{lyu2026matos,
title = {{MatOS: An Agent Operating System for Materials Research}},
author = {Dingyang Lyu and Jing Gao and Zhonghao Chang and Jingjie Huang and Xiangran Cheng and Xin He and Yongda Huang and Zefeng Li and Jiahui Liu and Jian Wen and Feng Yu and Bonan Zhu and Yongheng Li and Hongwei Du and Baole Wei and Ziheng Lu},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.34203},
eprint = {2610.34203},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.34203},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - MatOS: An Agent Operating System for Materials Research
AU - Dingyang Lyu
AU - Jing Gao
AU - Zhonghao Chang
AU - Jingjie Huang
AU - Xiangran Cheng
AU - Xin He
AU - Yongda Huang
AU - Zefeng Li
AU - Jiahui Liu
AU - Jian Wen
AU - Feng Yu
AU - Bonan Zhu
AU - Yongheng Li
AU - Hongwei Du
AU - Baole Wei
AU - Ziheng Lu
PY - 2026
DA - 2026/10/25
PB - arXiv
DO - 10.48550/arxiv.2610.34203
UR - https://arxiv.org/abs/2610.34203
AB - Materials research relies on scattered and growing data, knowledge, and simulation tools, and a single study often spans multi-day workflows across literature analysis, computation, and experiment. Large language model agents can now retrieve materials data, operate simulation codes, and close design-synthesis-characterization loops with autonomous laboratories. However, extending these agents with new capabilities, keeping long-running tasks on track, and letting them learn from past tasks and user feedback remain difficult. Here we present MatOS, an agent operating system for materials research with three core components: (1) a layered scientific asset graph that organizes data, knowledge, tools, skills, roles, and workflows as reusable assets; (2) a stateful adaptive workflow that runs tasks as explicit execution graphs with scientific checks, so that they can be resumed and traced; and (3) evidence-driven asset evolution, which revises these assets from execution records and user feedback. On ResearchClawBench, MatOS improves on the previously best-performing agent by 11.15% with the same base model and extends the cost-score Pareto frontier. With a common base model, it outperforms Codex CLI and Claude Code, exceeding Codex CLI by 2.13% on MatTools and 4.17% on SciAgentGym. In eight research cases, MatOS narrowed more than 30 million candidate structures to 97 solid-electrolyte candidates, identified 89 halide variants with higher computed conductivity than their parent structures, linked the force errors of 11 pretrained interatomic potentials to local atomic environments, and supported two new methods that reduce errors by 3.1% and 13.6% relative to their baselines. These results show that organizing research capabilities as reusable, evolving assets enables agents to carry out long, multistage materials studies. Code is available at https://github.com/DingyangLyu/MatOS.
ER -