eda flow automation

本文探讨了大型语言模型在电子设计自动化中的角色转变，从生成器到协调器，并提出需要一个标准化、物理感知的协调器来解决现有方法难以扩展到工业设计的问题。

To address strong script dependencies, high tool coupling in RTL-to-GDSII flows, the need for costly LLM fine-tuning, and the absence of a unified evaluation framework, this paper proposes a microservice-based LLM agent architecture. First, it introduces the Model Context Protocol to enable parallel context learning. Second, it replaces fine-tuning with structured prompt engineering to support natural-language-driven task decomposition and TCL parameter extraction. Third, it extends the CodeBLEU metric to quantitatively assess generated script quality. Evaluated on five EDA benchmarks, the approach achieves significant improvements: +28.6% average automation accuracy and 3.2× average speedup in execution time; generated scripts exhibit higher functional correctness and maintainability than baselines. The implementation is open-sourced to ensure reproducibility and foster community collaboration.

This work addresses the frequent failures of electronic design automation (EDA) code generated by large language models (LLMs), which often arise from violations of implicit structural dependencies among design entities—such as invalid paths, missing preconditions, or API incompatibilities. To overcome the high latency and poor scalability of existing tool-in-the-loop debugging approaches, the authors propose a novel framework for reliable code generation that operates without runtime feedback. The key innovation lies in explicitly modeling structural dependencies as execution contracts and guiding a validator-driven synthesis process via a structural dependency graph. This approach integrates graph-conditioned retrieval, constraint generation, and staged pre-execution validation. Empirical results demonstrate a single-step task pass rate of 82.5%, an improvement in multi-step task success from 30.0% to 84.0%, over twofold reduction in tool invocations, and a validator precision of 93.3% (6.7% false positive rate).

Large language models (LLMs) face significant challenges in automating EDA workflows—including insufficient tool understanding, highly heterogeneous interfaces, and error-prone long-chain tool invocations—resulting in low robustness and success rates. To address these issues, we propose EDAid, a multi-agent collaborative system specifically designed for EDA automation. EDAid introduces the novel “divergent reasoning, convergent goal” collaboration paradigm, wherein multiple specialized ChipLlama agents—each pursuing distinct reasoning paths—cooperatively execute cross-platform, multi-step tool chains. The system incorporates an API abstraction and adaptation layer, coupled with end-to-end toolchain validation and backtracking mechanisms, thereby overcoming error accumulation inherent to single-agent approaches under interface heterogeneity and long-range dependencies. Experiments demonstrate that EDAid reduces error rate by 62% over single-agent baselines, achieves a 98.3% workflow completion rate, enables seamless integration across Cadence, Synopsys, and Mentor platforms, and attains state-of-the-art performance on complex EDA tasks.

This work addresses the lack of systematic evaluation of AI agents across the full RTL-to-GDS electronic design automation (EDA) flow. To this end, we introduce FluxBench, a unified end-to-end benchmarking framework that standardizes prompt templates, tool environments, and process libraries, covering critical tasks including RTL generation, logic synthesis, placement and routing, and ECO automation. We propose Token ROI as a novel metric to quantify agent cost-effectiveness and demonstrate that agent architecture exerts a far greater influence on performance than the underlying foundation model. Experiments reveal that, with identical base models, different architectures exhibit performance gaps of up to 86.27% and Token ROI differences exceeding 105.92×. On the PicoRV32 benchmark, our proposed FluxEDA achieves an end-to-end score of 97.94, representing an 8.39× improvement over Claude Code.

This work addresses the frequent breakdowns in handoffs across tool, session, and organizational boundaries within electronic design automation (EDA) workflows, which often result in design artifacts failing to meet explicit or implicit downstream requirements. To tackle this challenge, the paper introduces the core principle of “handoff validity,” defines three boundary system types—Stage-, Flow-, and Organization-Bound—and proposes a five-layer EDA Agent Communication Protocol (EACP) to enable trustworthy transmission of context, evidence, and provenance information. Leveraging large language model–based agents, the framework supports reliable collaboration through capabilities such as executable script generation, knowledge retrieval, state management, and secure intellectual property protocols. The study also presents a systematic survey of 82 related systems, establishes a unified terminology, and outlines a research agenda to lay the foundation for intelligent, cross-boundary EDA.

This work proposes a unified, stateful agent-based execution infrastructure for integrating large language models with electronic design automation (EDA) tools. Addressing the limitations of existing stateless scripting or request-response paradigms—which hinder iterative optimization in production environments—the architecture introduces a managed gateway interface coupled with persistent EDA tool instances. This enables, for the first time, an execution layer that supports state reuse, rollback, and collaborative iteration. By employing a structured request/response mechanism, the system facilitates state-preserving interactions with heterogeneous EDA tools. The approach is validated through two commercial use cases: post-routing timing engineering change orders (ECOs) and standard cell sublibrary optimization, where it successfully orchestrates multi-step analysis and refinement workflows, demonstrating its effectiveness and practicality in agent-driven EDA automation.

This work addresses the growing bottleneck in front-end chip design caused by escalating circuit complexity and compressed time-to-market constraints. To overcome this challenge, the paper proposes a novel paradigm that integrates large language models (LLMs) with AI agents to establish a unified intelligent interface capable of automatically generating hardware description language (HDL) code, constructing testbenches, and exploring the design space. Building upon agent-based architectures such as OpenClaw, the project systematically advances front-end EDA toward greater autonomy and intelligence. The approach demonstrates notable progress in co-generating circuits and testbenches while optimizing design quality. Furthermore, the study clarifies key technical challenges and outlines promising directions for future research in intelligent EDA methodologies.

This study addresses the challenge faced by production system engineers in automatically verifying production line layouts due to limited knowledge of PDDL and planning theory. To bridge this gap, the authors propose a novel approach based on an Asset Administration Shell (AAS) capability model that natively generates complete PDDL planning problems directly from domain-level descriptions, eliminating the need for PDDL-specific submodels. The method integrates four Industry 4.0 standards—VDI 3682, IEC 61360-1, IDTA 02011, and IDTA 02016—to construct the AAS and employs an extraction algorithm to automatically translate multi-AAS architectures into PDDL domains. In a laboratory case study, the approach enabled engineers to systematically compare four layout variants by modifying only the AAS model, significantly lowering the barrier to adopting automated planning in industrial settings.

本文提出了一种基于大型语言模型的行为驱动硬件开发流程，通过定义形式验证Gherkin场景来减少自然语言规范的模糊性，提高硬件设计的形式验证效果。