Product

Gauge

Optimal thresholds for quantum error correcting codes.

What is Gauge

Gauge is a Python package that computes the optimal error threshold for any quantum error correcting code and noise model. It is part of the Plaquette family of tools, extending Plaquette’s architecture simulation capabilities with theoretical threshold analysis.

The threshold Gauge computes is not the threshold of a specific decoder. It is the theoretical ceiling: the maximum error rate below which any decoder could, in principle, drive the logical error rate to zero by scaling up the code. Gauge computes this ceiling by mapping the decoding problem onto a statistical mechanical spin model, then using Monte Carlo simulation to find the phase transition that marks the boundary between correctable and uncorrectable regimes.

How it works

Gauge maps the decoding problem onto a statistical mechanical spin model.

Phase diagram showing the ordered and disordered phases of the spin model, with the phase boundary intersecting the Nishimori line at the optimal error threshold p_opt.
The threshold is the intersection of the phase boundary with the Nishimori line.

The physical error probabilities are wired into the model’s couplings through the Nishimori conditions, which ensure the partition function directly gives the probability of each error class. The optimal threshold is the phase transition in this model: where the phase boundary intersects the Nishimori line.

Who uses Gauge

Quantum hardware teams, QEC researchers, decoder developers, and anyone else working on fault-tolerance development.

What they use it for

Gauge tells you the best any decoder could do on your code and noise model, answering the question: is the gap between your current decoder and the optimal threshold worth closing with a better decoder, or do the code or hardware need to be improved? Gauge gives you that answer before you commit resources to either direction.

It supports a growing range of quantum error correction codes and noise models. New code families are added regularly, and anything not yet in the built-in library can be defined directly from stabilizers. The Monte Carlo engine runs 100 times faster than those reported in the literature, making it practical for real architectural studies rather than toy examples.

Gauge vs. Plaquette

Gauge and Plaquette are complementary tools. Plaquette tells you the threshold your decoder achieves today. Gauge tells you the theoretical ceiling. Together, they show how much room is left and where to focus next.

Get started

If you’re building or scaling quantum hardware, get in touch to discuss licensing or book a demo.

Not ready yet? Join our practitioners mailing list to be notified when we release new features or publish technical content.