Bohrium + SciMaster: Building the infrastructure and ecosystem for agentic science at scale

Linfeng Zhang; Siheng Chen; Yuzhu Cai; Jingyi Chai; Junhan Chang; Kun Chen; Zhi X. Chen; Zhaohan Ding; Yuwen Du; Yuanpeng Gao; Yuan Gao; Jing Gao; Zhifeng Gao; Qiangqiang Gu; Yanhui Hong; Yuan Huang; Xi Fang; Xiaohong Ji; Guolin Ke; Zixing Lei; Xinyu Li; Yongge Li; Ruoxue Liao; Hang Lin; Xiaolu Lin; Fenyi Liu; Yuxiang Liu; Xinzijian Liu; Zexi Liu; Jintan Lu; Tingjia Miao; Haohui Que; Weijie Sun; Yanfeng Wang; Bingyang Wu; Tianju Xue; Rui Ye; Jinzhe Zeng; Duo Zhang; Jiahui Zhang; Linfeng Zhang; Tianhan Zhang; Wenchang Zhang; Yuzhi Zhang; Zezhong Zhang; Hang Zheng; Hui Zhou; Tong Zhu; Xinyu Zhu; Qingguo Zhou; Weinan E

doi:10.59717/ipj.aiplus.2026.100005

AI Plus > 2026 Vol. 1 > No. 1 > 100005

Next Article Previous Article

ARTICLE Open Access

Cite PDF

Bohrium + SciMaster: Building the infrastructure and ecosystem for agentic science at scale

1.
DP Technology, Beijing 100080, China
2.
AI for Science Institute, Beijing 100084, China
3.
School of Artificial Intelligence, Shanghai Jiao Tong University, Shanghai 200240, China
4.
Beihang University, Beijing 100191, China
5.
Peking University, Beijing 100871, China
6.
Institute of Theoretical Physics, Chinese Academy of Sciences, Beijing 100190, China
7.
Shanghai Innovation Institute, Shanghai 200231, China
8.
East China Normal University, Shanghai 200062, China
9.
Zhongguancun Academy, Beijing 100094, China
10.
University of Science and Technology of China, Hefei 230026, China
11.
Tongji University, Shanghai 200092, China
12.
The Hong Kong University of Science and Technology, Hong Kong SAR 999077, China

Show all affliationsShow less

More Information

Corresponding author: sihengc@sjtu.edu.cn (S.C.)
Received Date: July 21, 2026

Accepted Date: September 20, 2026

Published Online: September 24, 2026

AI Plus 1(1), https://doi.org/10.59717/ipj.aiplus.2026.100005 | Cite this article

GRAPHICAL ABSTRACT

GRAPHICAL ABSTRACT

DownLoad: Full size image

PUBLIC SUMMARY

PUBLIC SUMMARY
1. Bohrium makes scientific assets callable, observable, and governed for agentic workflows.
  
  SciMaster coordinates long-horizon workflows across Reading, Computing, and Experiment.
  
  Eleven master agents report faster cycles in search, simulation, optimization, and design.
  
  The stack points toward Science-as-a-Service with validation and human judgment.

ABSTRACT

ABSTRACT

AI agents are emerging as a practical way to run multi-step scientific workflows that interleave reasoning, tool use, and verification. Scaling such agentic science remains difficult because workflows are hard to observe and reproduce, many scientific tools and laboratory systems are not agent-ready, and execution traces are often fragmented. We present Bohrium+SciMaster as an infrastructure-and-ecosystem approach to this problem. Bohrium turns scientific data, software, compute, and laboratory systems into governed, traceable, agent-ready capabilities, while SciMaster orchestrates these capabilities into long-horizon scientific workflows. Between them, a scientific intelligence substrate organizes reusable models, knowledge, and components into executable building blocks for workflow reasoning and action. Across eleven representative master agents spanning literature analysis, simulation, optimization, materials design, and experiment-facing workflows, we report deployment-observed reductions in end-to-end scientific cycle time, together with execution-grounded signals from real workloads at multi-million scale. Together, these results point toward Science-as-a-Service: scientific workflows become increasingly executable, observable, and continuously improvable while remaining anchored in traceable validation and human judgment.
- agentic science at scale /
- scientific production /
- agent-ready infrastructure /
- scientific workflows /
- Science-as-a-Service

FullText(HTML)

REFERENCES

[1]	Jumper J., Evans R., Pritzel A. et al. (2021). Highly accurate protein structure prediction with AlphaFold. Nature 596:583−589. DOI:10.1038/s41586-021-03819-2 View in Article CrossRef Google Scholar
[2]	Merchant A., Batzner S., Schoenholz S. S. et al. (2023). Scaling deep learning for materials discovery. Nature 624:80−85. DOI:10.1038/s41586-023-06735-9 View in Article CrossRef Google Scholar
[3]	Zhang L., Han J., Wang H. et al. (2018). Deep potential molecular dynamics: A scalable model with the accuracy of quantum mechanics. Phys. Rev. Lett. 120:143001. DOI:10.1103/PhysRevLett.120.143001 View in Article CrossRef Google Scholar
[4]	Mitchener L., Yiu A., Chang B. et al. (2025). Kosmos: An AI scientist for autonomous discovery. arXiv:2511.02824. View in Article Google Scholar
[5]	Ghareeb A. E., Chang B., Mitchener L. et al. (2026). A multi-agent system for automating scientific discovery. Nature 655:497−505. DOI:10.1038/s41586-026-10652-y View in Article CrossRef Google Scholar
[6]	Gottweis J., Weng W.-H., Daryin A. et al. (2026). Accelerating scientific discovery with Co-Scientist. Nature 655:487−496. DOI:10.1038/s41586-026-10644-y View in Article CrossRef Google Scholar
[7]	Wu M., Wang Y., Ming Y. et al. (2025). CheMatAgent: Enhancing LLMs for chemistry and materials science through tree-search based tool learning. arXiv:2506.07551. View in Article Google Scholar
[8]	Ramos M. C., Collison C. J. and White A. D. (2025). A review of large language models and autonomous agents in chemistry. Chem. Sci. 16:2514−2572. DOI:10.1039/D4SC03921A View in Article CrossRef Google Scholar
[9]	Chai J., Tang S., Ye R. et al. (2025). SciMaster: Towards general-purpose scientific AI agents, Part I. X-Master as foundation: Can we lead on humanity's last exam? arXiv:2507.05241. View in Article Google Scholar
[10]	Sim M., Vakili M. G., Strieth-Kalthoff F. et al. (2024). ChemOS 2.0: An orchestration architecture for chemical self-driving laboratories. Matter 7:2959−2977. DOI:10.1016/j.matt.2024.04.022 View in Article CrossRef Google Scholar
[11]	Gao J., Chang J., Que H. et al. (2025). UniLabOS: An AI-native operating system for autonomous laboratories. arXiv:2512.21766. View in Article Google Scholar
[12]	Li Y., Huang Y., Wang T. et al. (2025). Inverse knowledge search over verifiable reasoning: Synthesizing a scientific encyclopedia from a long chains-of-thought knowledge base. arXiv:2510.26854. View in Article Google Scholar
[13]	Zhang D., Peng A., Cai C. et al. (2026). A graph neural network for the era of large atomistic models. npj Comput. Mater. 12:276. DOI:10.1038/s41524-026-02146-2 View in Article CrossRef Google Scholar
[14]	Zhou G., Gao Z., Ding Q. et al. (2023). Uni-Mol: A universal 3D molecular representation learning framework. In: The Eleventh International Conference on Learning Representations (ICLR). View in Article Google Scholar
[15]	Wang X., Gu R., Chen Z. et al. (2023). UNI-RNA: Universal pre-trained models revolutionize RNA research. bioRxiv. DOI: 10.1101/2023.07.11.548588 View in Article Google Scholar
[16]	Fang X., Hong Y., Wang N. et al. (2025). MolParser: End-to-end visual recognition of molecule structures in the wild. In: Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), pp. 24528–24538. DOI: 10.1109/ICCV51701.2025.02274 View in Article Google Scholar
[17]	Gao Z., Ji X., Zhao G. et al. (2023). Uni-QSAR: An auto-ML tool for molecular property prediction. arXiv:2304.12239. View in Article Google Scholar
[18]	Hong Y., Wang N., Xia Z. et al. (2025). Uni-AIMS: AI-powered microscopy image analysis. arXiv:2505.06918. View in Article Google Scholar
[19]	Wen Z., Yang B., Chen S. et al. (2026). Innovator-VL: A multimodal large language model for scientific discovery. arXiv:2601.19325. View in Article Google Scholar
[20]	Wang H., Zhang L., Han J. et al. (2018). DeePMD-kit: A deep learning package for many-body potential energy representation and molecular dynamics. Comput. Phys. Commun. 228:178−184. DOI:10.1016/j.cpc.2018.03.016 View in Article CrossRef Google Scholar
[21]	Zhou W., Zheng D., Liu Q. et al. (2025). ABACUS: An electronic structure analysis package for the AI era. J. Chem. Phys. 163:192501. DOI:10.1063/5.0297563 View in Article CrossRef Google Scholar
[22]	Gu Q., Zhouyin Z., Pandey S. K. et al. (2024). Deep learning tight-binding approach for large-scale electronic simulations at finite temperatures with ab initio accuracy. Nat. Commun. 15:6772. DOI:10.1038/s41467-024-51006-4 View in Article CrossRef Google Scholar
[23]	Zhouyin Z., Gan Z., Pandey S. K. et al. (2025). Learning local equivariant representations for quantum operators. In: The Thirteenth International Conference on Learning Representations (ICLR). View in Article Google Scholar
[24]	Zou J., Zhouyin Z., Lin D. et al. (2025). Deep learning accelerated quantum transport simulations in nanoelectronics: From break junctions to field-effect transistors. npj Comput. Mater. 11:375. DOI:10.1038/s41524-025-01853-6 View in Article CrossRef Google Scholar
[25]	Mao R., Lin M., Zhang Y. et al. (2023). DeepFlame: A deep learning empowered open-source platform for reacting flow simulations. Comput. Phys. Commun. 291:108842. DOI:10.1016/j.cpc.2023.108842 View in Article CrossRef Google Scholar
[26]	Xue T., Liao S., Gan Z. et al. (2023). JAX-FEM: A differentiable GPU-accelerated 3D finite element solver for automatic inverse design and mechanistic data science. Comput. Phys. Commun. 291:108802. DOI:10.1016/j.cpc.2023.108802 View in Article CrossRef Google Scholar
[27]	Zhang Y., Wang H., Chen W. et al. (2020). DP-GEN: A concurrent learning platform for the generation of reliable deep learning based potential energy models. Comput. Phys. Commun. 253:107206. DOI:10.1016/j.cpc.2020.107206 View in Article CrossRef Google Scholar
[28]	Liu X., Han Y., Li Z. et al. (2024). Dflow: A Python framework for constructing cloud-native AI-for-Science workflows. arXiv:2404.18392. View in Article Google Scholar
[29]	Yuan F., Ding Z., Liu Y.-P. et al. (2025). DPDispatcher: Scalable HPC task scheduling for AI-driven science. J. Chem. Inf. Model. 65:12155−12160. DOI:10.1021/acs.jcim.5c02081 View in Article CrossRef Google Scholar
[30]	Zeng J., Peng X., Zhuang Y.-B. et al. (2025). dpdata: A scalable Python toolkit for atomistic machine learning data sets. J. Chem. Inf. Model. 65:11497−11504. DOI:10.1021/acs.jcim.5c01767 View in Article CrossRef Google Scholar
[31]	Li Z., Wen T., Zhang Y. et al. (2025). APEX: An automated cloud-native material property explorer. npj Comput. Mater. 11:88. DOI:10.1038/s41524-025-01580-y View in Article CrossRef Google Scholar
[32]	Taniai T., Igarashi R., Suzuki Y. et al. (2024). Crystalformer: Infinitely connected attention for periodic structure encoding. In: The Twelfth International Conference on Learning Representations (ICLR). View in Article Google Scholar
[33]	Liu Z., Cai Y., Zhu X. et al. (2025). ML-Master: Towards AI-for-AI via integration of exploration and reasoning. arXiv:2506.16499. View in Article Google Scholar
[34]	Fan E., Hu K., Wu Z. et al. (2026). ChatCFD: A large language model-driven agent for end-to-end computational fluid dynamics automation with structured knowledge and reasoning. Adv. Intell. Discovery 2:e202500174. DOI:10.1002/aidi.202500174 View in Article CrossRef Google Scholar

Cited by

ABOUT THIS ARTICLE

Cite this article:

Zhang L., Chen S., Cai Y., et al. (2026). Bohrium + SciMaster: Building the infrastructure and ecosystem for agentic science at scale. AI Plus 1:100005. https://doi.org/10.59717/ipj.aiplus.2026.100005

Zhang L., Chen S., Cai Y., et al. (2026). Bohrium + SciMaster: Building the infrastructure and ecosystem for agentic science at scale. AI Plus 1:100005. https://doi.org/10.59717/ipj.aiplus.2026.100005

Welcome!

To request copyright permission to republish or share portions of our works, please visit Copyright Clearance Center's (CCC) Marketplace website at marketplace.copyright.com.

PDF

Figures(4) Tables(3)

Supplementary Information

download_supplementary
SI-Extended-AI-DOI-5

Article Metrics

Article views(86) PDF downloads(23)

Bohrium + SciMaster: Building the infrastructure and ecosystem for agentic science at scale

GRAPHICAL ABSTRACT

PUBLIC SUMMARY

ABSTRACT