DeepCircuitX: A Comprehensive Repository-Level Dataset for RTL Code Understanding, Generation, and PPA Analysis

TL;DR

DeepCircuitX is a comprehensive, multi-level RTL dataset with detailed annotations, netlists, and PPA metrics, designed to improve machine learning models for RTL understanding, generation, and PPA prediction.

Contribution

It introduces a holistic, repository-level RTL dataset with Chain of Thought annotations, enabling advanced training and evaluation of large language models for hardware design tasks.

Findings

Effective fine-tuning of LLMs on DeepCircuitX improves RTL understanding and generation.

The dataset enables accurate PPA prediction directly from RTL code.

Human evaluations confirm the dataset's high quality and utility.

Abstract

This paper introduces DeepCircuitX, a comprehensive repository-level dataset designed to advance RTL (Register Transfer Level) code understanding, generation, and power-performance-area (PPA) analysis. Unlike existing datasets that are limited to either file-level RTL code or physical layout data, DeepCircuitX provides a holistic, multilevel resource that spans repository, file, module, and block-level RTL code. This structure enables more nuanced training and evaluation of large language models (LLMs) for RTL-specific tasks. DeepCircuitX is enriched with Chain of Thought (CoT) annotations, offering detailed descriptions of functionality and structure at multiple levels. These annotations enhance its utility for a wide range of tasks, including RTL code understanding, generation, and completion. Additionally, the dataset includes synthesized netlists and PPA metrics, facilitating early-stage design exploration and enabling accurate PPA prediction directly from RTL code. We demonstrate the dataset's effectiveness on various LLMs finetuned with our dataset and confirm the quality with human evaluations. Our results highlight DeepCircuitX as a critical resource for advancing RTL-focused machine learning applications in hardware design automation.Our data is available at https://zeju.gitbook.io/lcm-team.