Skip to content

OpenROAD-Assistant/OpenROAD-Agent

Repository files navigation

OpenROAD Agent

This repository contains scripts that enable large-language models (LLMs) to serve as agents for interacting with OpenROAD and reproducing the results in the paper.

Framework Description

Inference Large language models (LLMs) are increasingly being used in various domains, including chip design. Recent works have demonstrated the effectiveness of LLMs in EDA tool script generation. However, these LLMs can often hallucinate API calls even when directly provided API documentation and context. As a result, this leads to significant errors and inefficiencies in chip design workflows. To address this, we introduce OpenROAD Agent, an LLM that integrates directly with OpenROAD, an open-source tool for physical design. OpenROAD Agent enables real-time script generation with error detection and correction. OpenROAD Agent autonomously generates and executes Python code within OpenROAD while dynamically refining outputs based on tool feedback. OpenROAD Agent leverages Qwen2.5-Coder and employs a hybrid supervised and reinforcement learning training to improve code correction capabilities, usability, and hallucination frequency. OpenROAD Agent achieves an accuracy of 94% on script generation tasks and outperforms both prior work and existing foundation models.

Model Weights

Model weights can be found on Google Drive. After downloading, please place the two folders in src/Saved_Model.

Table of Content

Build OpenROAD and Pythonize the Macro Placer

The following technique assumes you have a machine with the required Ubuntu OS prerequisite of OpenROAD.

Install dependencies for OpenROAD:

sudo ./OpenROAD/etc/DependencyInstaller.sh

Once dependencies have been installed, build OpenROAD and Pythonize the macro placer:

cd ./OpenROAD_update
bash replace.sh
cd ../OpenROAD/
mkdir build
cd build
cmake ..
make -j

Training and Testing:

Please see src.

Cite this work

B. -Y. Wu, U. Sharma, A. Rovinski, and V. A. Chhabria, "OpenROAD Agent: An Intelligent Self-Correcting Script Generator for OpenROAD," 2025 IEEE International Conference on LLM-Aided Design (ICLAD), Stanford, California (CA), USA, 2025,

About

OpenROAD Agent. This repository contain the model to train and testing the model using EDA Corpus dataset.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors