Welcome to SiliconCompiler’s Documentation!#

Version: 0.23.3

Useful Links: Installation | GitHub Repo | File an Issue

What is SiliconCompiler?#

SiliconCompiler is an open source, modular, build system that automates translation from hardware design source code to silicon (“make for silicon”).


Silicon had an enormous positive impact on the world over the last 50 years and it is a social imperative that we surf the exponential Moore’s Law as long as possible. Extreme hardware specialization is the only viable path for extending the current exponential electronics improvement trajectory indefinitely, but the path is currently blocked by the high engineering effort of chip design.

Hardware specialization for a long tail of future applications will require the creation of completely automated end-to-end compilers that are orders of magnitude faster than today’s tools. The enormity of these challenges means single machine execution and monolithic single company efforts are unlikely to be sufficient. In this work, we take a distributed systems approach to compilers, with the goal of creating infrastructure that scales to thousands of developers and millions of servers.

Our Approach#

The SiliconCompiler project is based on a standardized Schema that supports orthogonal combinations of design, tools, and Process Design Kits (PDKs). The schema design philosophy is to “make the complex possible while keeping the simple simple”.

To simplify flow development, the project incorporates a simple object oriented Python API. The API includes abstracted set/get access to the Schema, a flowgraph based parallel programming model, and a suite of utility functions for compilation setup and metric tracking.


The expansive data schema, standardized plug-in interfaces, and built-in dynamic module search functionality enables SiliconCompiler to scale effectively to a large number of tools and PDKs. The reference manual serves as a good starting point for folks who want to add their own PDKs and tools.

To further reduce design access barriers, the project also supports a client-server execution model that leverages the cloud to: 1) reduce tool installation barriers, 2) reduce the barrier to massively parallel elastic compute, and 3) address the NDA barrier for PDK and EDA tools.


Supported Technologies#




C, Verilog, SV, VHDL, Chisel, Migen/Amaranth, Bluespec


Verilator, Icarus, GHDL, Xyce


Yosys, Vivado, Synopsys, Cadence


OpenRoad, Synopsys, Cadence


VPR, nextpnr, Vivado

Layout Viewer

Klayout, OpenRoad, Cadence, Synopsys


Magic, Synopsys, Siemens


sky130, gf180, asap7, freepdk45, gf12lp, intel16


SiliconCompiler project authors in chronological order: Andreas Olofsson, William Ransohoff, Noah Moroze, Zachary Yedidia, Massimiliano Giacometti, Kimia Talaei, Peter Gadfort, Aulihan Teng, Peter Grossmann, Gabriel Aguirre, Martin Troiber.

Getting started#

The fastest way to get started, after installation, is to walk through a simple demo!

A more complete introduction for new users, the Introductory User Guide section is intended for those who want to use their own designs using pre-defined build flows.

Reference Manual section contains useful lookup information, like flow building blocks, the schema and other API references.

Advanced Guide is intended for users who are already familiar with the basics in User Guide and would like to build their own flows or add additional functionality.


  • Search Glossary,

  • Review FAQ, or

  • File an issue: if you can’t find something, or something is not working, the SiliconCompiler team is happy to help!