CapBench: A Multi-PDK Dataset for Machine-Learning-Based Post-Layout Capacitance Extraction

TL;DR

CapBench provides a comprehensive, multi-technology dataset for machine learning-based capacitance extraction, enabling transfer learning, scalability studies, and benchmarking of various ML architectures.

Contribution

This work introduces a reproducible multi-PDK dataset with high-fidelity labels, diverse design types, and baseline ML models, advancing research in post-layout capacitance extraction.

Findings

CNNs achieve 1.75% error in capacitance prediction.

GNNs are up to 41.4x faster but have 10.2% error.

The dataset supports transfer learning and scalability studies.

Abstract

We present CapBench, a fully reproducible, multi-PDK dataset for capacitance extraction. The dataset is derived from open-source designs, including single-core CPUs, systems-on-chip, and media accelerators. All designs are fully placed and routed using 14 independent OpenROAD flow runs spanning three technology nodes: ASAP7, NanGate45, and Sky130HD. From these layouts, we extract 61,855 3D windows across three size tiers to enable transfer learning and scalability studies. High-fidelity capacitance labels are generated using RWCap, a state-of-the-art random-walk solver, and validated against the industry-standard Raphael, achieving a mean absolute error of 0.64% for total capacitance. Each window is pre-processed into density maps, graph representations, and point clouds. We evaluate 10 machine learning architectures that illustrate dataset usage and serve as baselines, including convolutional neural networks (CNNs), point cloud transformers, and graph neural networks (GNNs). CNNs demonstrate the lowest errors (1.75%), while GNNs are up to 41.4x faster but exhibit larger errors (10.2%), illustrating a clear accuracy-speed trade-off. Code and dataset are available at https://github.com/THU-numbda/CapBench.