Can AI Agents Really Complete RTL-to-GDS? Lessons from Benchmarking Tool-Interactive EDA Workflows

2026-07-20

AI Summary

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.

Abstract

LLM-driven agent systems have emerged as a promising paradigm for electronic design automation (EDA), demonstrating strong potential for automating complex design workflows. However, existing evaluations primarily examine individual language models on isolated EDA tasks, providing limited insight into how different agent systems perform across complete EDA flows. In this work, we present FluxBench, a systematic evaluation of AI agents on end-to-end EDA workflows under unified prompts, tool environments, and technology library settings. Our evaluation covers representative scenarios, including RTL generation with open-source toolchains and an RTL-to-GDS flow using closed-source commercial EDA tools for industrial applications. Through these workflows, we assess agents' capabilities in RTL code generation, iterative repair, tool-feedback utilization, logic synthesis, placement and routing (P&R), and Engineering Change Order (ECO) automation. To further characterize the efficiency of agent systems, we introduce Token ROI, a cost-efficiency metric that measures effective improvements in EDA artifacts relative to token usage and runtime cost. Experimental results show that, even when built on the same foundation model, different agent system architectures can exhibit performance gaps of up to 86.27%. Moreover, among systems with comparable task performance, Token ROI can differ by as much as $105.92\times$. In the RTL-to-GDS flow using PicoRV32 as a case study, FluxEDA achieves an end-to-end score of up to 97.94, outperforming Claude Code equipped with domain-specific EDA skills by up to $8.39\times$. These results indicate that domain-specific skills alone are insufficient to improve agent performance in large-scale EDA scenarios. Instead, both agent system design and foundation model capability play critical roles in enabling effective automated EDA workflows.

Problem

• AI Agents
• Electronic Design Automation
• RTL-to-GDS
• Benchmarking
• End-to-End Workflow

Innovation

• FluxBench
• RTL-to-GDS
• AI Agents
• Token ROI
• EDA Automation