收藏切换
SimulatorGen: An LLM-Based Multi-Agent Framework for Automatic Generation of DNN Accelerator Simulators
收藏切换
Yuhuan Xia, Tun Li*, Xianfa Zhou, Wenbo Zhao, Ruiyu Zhang, Yang Guo
Journal of Computer Research and Development | 2026, 63(6) : 1392 - 1408
Less
收藏切换
Journal of Computer Research and Development | 2026, 63(6): 1392-1408
• Special Issue on LLMs Aid Chip Design •
SimulatorGen: An LLM-Based Multi-Agent Framework for Automatic Generation of DNN Accelerator Simulators
Full
Yuhuan Xia, Tun Li*, Xianfa Zhou, Wenbo Zhao, Ruiyu Zhang, Yang Guo
Affiliations
  • College of Computer, National University of Defense Technology, Changsha 410073
  • born in 2000. PhD candidate. Her main research interest includes DNN accelerator simulators

    born in 1974. Professor, PhD supervisor. Senior member of CCF. His main research interest includes electronic design automation

    born in 2001. Master candidate. His main research interest includes the mapping of neural networks on in-memory accelerators

    born in 1999. PhD candidate. His main research interest includes fuzz testing in hardware development

    born in 2000. PhD candidate. Her main research interest includes electronic design automation

    born in 1971. Professor, PhD supervisor. Senior member of CCF. His main research interests include very-large-scale inte-gration (VLSI) and processor architecture

About Author:

born in 2000. PhD candidate. Her main research interest includes DNN accelerator simulators

born in 1974. Professor, PhD supervisor. Senior member of CCF. His main research interest includes electronic design automation

born in 2001. Master candidate. His main research interest includes the mapping of neural networks on in-memory accelerators

born in 1999. PhD candidate. His main research interest includes fuzz testing in hardware development

born in 2000. PhD candidate. Her main research interest includes electronic design automation

born in 1971. Professor, PhD supervisor. Senior member of CCF. His main research interests include very-large-scale inte-gration (VLSI) and processor architecture

Published: 2026-06-01 doi: 10.7544/issn1000-1239.202660116
Outline
收藏切换

With the rapid development of deep neural network (DNN) accelerators, building simulators for new architectures is costly and time-consuming. Although advances in large language models (LLMs) have opened possibilities for automated simulator generation, existing approaches suffer from limited generality, inability to construct complete systems, and high construction complexity. To address these challenges, we propose SimulatorGen, a multi-agent framework that generates DNN accelerator simulator code from natural language descriptions. First, we abstract the architecture of DNN accelerator simulators and extract twenty-three component specifications. Based on this abstraction, four collaborative agents are introduced to accomplish generation: the analyst agent retrieves domain knowledge from the simulator library via retrieval-augmented generation (RAG) and constructs structured prompts by leveraging chain-of-thought (CoT) reasoning; the coder agent generates or refines code using prompts and test feedback; the tester agent performs syntax checking, functional testing, and formal verification using the Z3 solver based on properties extracted from specifications; and the assembly agent conducts component integration, automated execution, and metric comparison to enable end-to-end construction. We evaluate SimulatorGen on twenty-three generation tasks covering diverse DNN accelerator modules and architectures. Experimental results show that SimulatorGen built on GPT-4o outperforms LLM baselines, including Claude-Sonnet-4, achieving a Pass@1 score of 82.39%. Furthermore, using the successfully generated components, we construct runnable simulators for tensor processing unit (TPU) and MAERI architectures. Compared with STONNE, the simulators built by SimulatorGen achieve relative errors ranging from 1.31% to 7.34% in energy, latency, and energy-delay product (EDP) across multiple DNN models, while maintaining functional consistency verified through testing and execution, demonstrating faithful modeling of accelerator behavior. In contrast to the single-agent SimulatorCoder, which only supports module replacement, SimulatorGen enables end-to-end generation of complete simulators, further validating the effectiveness of the proposed approach.

DNN accelerator simulator  /  large language model  /  multi-agent  /  code generation  /  retrieval-augmented generation (RAG)
夏雨欢, 李暾, 周贤发, 赵文博, 张瑞瑜, 郭阳. SimulatorGen:基于大语言模型多智能体的DNN加速器模拟器自动生成框架. 计算机研究与发展, 2026 , 63 (6) : 1392 -1408 . DOI: 10.7544/issn1000-1239.202660116
Yuhuan Xia, Tun Li, Xianfa Zhou, Wenbo Zhao, Ruiyu Zhang, Yang Guo. SimulatorGen: An LLM-Based Multi-Agent Framework for Automatic Generation of DNN Accelerator Simulators[J]. Journal of Computer Research and Development, 2026 , 63 (6) : 1392 -1408 . DOI: 10.7544/issn1000-1239.202660116
Year 2026 volume 63 Issue 6
83
38
Cite this Article
BibTeX
Article Info
doi: 10.7544/issn1000-1239.202660116
  • Receive Date:2026-03-12
  • Online Date:2026-11-17
  • Published:2026-06-01
Article Data
Affiliations
History
  • Received:2026-03-12
  • Revised:2026-05-06
Affiliations
    College of Computer, National University of Defense Technology, Changsha 410073
References
Share
https://castjournals.cast.org.cn/joweb/jsjyjyfz/EN/10.7544/issn1000-1239.202660116
Share to
QR

Scan QR to access full text

Cite this article
BibTeX
Citations
表12种不同金属材料的力学参数
科
Family
属数
Number of
genus
种数
Number of
species
占总种数比例
Percentage of
total species (%)
属
Genus
种数
Number of
species
占总种数比例
Percentage of total
species (%)
鹅膏菌科Amanitaceae 2 11 5.26 鹅膏菌属 Amanita 10 4.78
小菇科 Mycenaceae 2 12 5.74 丝盖伞属 Inocybe 5 2.39
多孔菌科 Polyporaceae 8 14 6.70 蜡蘑属 Laccaria 5 2.39
红菇科 Russulaceae 3 23 11.00 小皮伞属 Marasmius 6 2.87
小菇属 Mycena 11 5.26
光柄菇属 Pluteus 5 2.39
红菇属 Russula 17 8.13
栓菌属 Trametes 5 2.39
关闭全屏
  • BibTeX
  • EndNote
  • RefWorks
  • TxT