HDL-GPT: High-Quality HDL is All You Need

TL;DR

HDL-GPT introduces a novel approach leveraging high-quality open-source HDL data to train large models with exceptional performance and zero-shot generalization, surpassing current state-of-the-art in various circuit design tasks.

Contribution

The paper demonstrates that high-quality HDL data alone can produce superior models, highlighting data curation and augmentation as key to model performance.

Findings

HDL-GPT outperforms SOTA HDL models by 50-200% on benchmarks.

Careful data filtering and augmentation improve model quality.

Different fine-tuning methods significantly impact results.

Abstract

This paper presents Hardware Description Language Generative Pre-trained Transformers (HDL-GPT), a novel approach that leverages the vast repository of open-source High Definition Language (HDL) codes to train superior quality large code models. The core premise of this paper is the hypothesis that high-quality HDL is all you need to create models with exceptional performance and broad zero-shot generalization abilities. The paper elucidates the methods employed for the curation and augmentation of large corpora from open-source HDL code, transforming highly variable quality data into high-quality data through careful prompting and context maintenance. We demonstrate that the careful selection, filtering, and augmentation of data across HDLs can yield powerful models that surpass current state-of-the-art models. We also explore the impact of different fine-tuning methods on the quality of results. We describe experimental results across a range of fine-tuned SOTA LLMs, substantiating our claims. We demonstrate improvements of 50% to 200% over SOTA HDL models on current benchmarks in tasks ranging from HDL circuit explanations, code generation, formal and simulation testbench creation, triaging bugs, and fixing them. HDL-GPT opens new avenues for the development of advanced model training techniques for circuit design tasks.