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

arXiv: https://arxiv.org/abs/2604.11202 · HTML: https://mlsanity.com/paper/2604.11202/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 8 of its 25 references that are in the map.
- First posted on arXiv 13 Apr 2026.

## Builds on (most cited first, 8 of 8)

- [Graph Attention Networks](https://mlsanity.com/paper/1710.10903/) (2017)
- [OpenABC-D: A Large-Scale Dataset For Machine Learning Guided Integrated Circuit Synthesis](https://mlsanity.com/paper/2110.11292/) (2021)
- [Benchmarking End-To-End Performance of AI-Based Chip Placement Algorithms](https://mlsanity.com/paper/2407.15026/) (2024)
- [Machine Learning Based Routing Congestion Prediction in FPGA High-Level Synthesis](https://mlsanity.com/paper/1905.03852/) (2019)
- [ForgeEDA: A Comprehensive Multimodal Dataset for Advancing EDA](https://mlsanity.com/paper/2505.02016/) (2025)
- [A Thermal Machine Learning Solver For Chip Simulation](https://mlsanity.com/paper/2209.04741/) (2022)
- [CNN-Cap: Effective Convolutional Neural Network Based Capacitance Models for Full-Chip Parasitic Extraction](https://mlsanity.com/paper/2107.06511/) (2021)
- [Efficient FRW Transitions via Stochastic Finite Differences for Handling Non-Stratified Dielectrics](https://mlsanity.com/paper/2507.09730/) (2025)

## Built on by (most cited first, 1 of 1)

- [Flash-CNNCap: Capacitance Extraction via Image Mapping](https://mlsanity.com/paper/2607.23877/) (2026)

## Abstract (arXiv)

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.