Exponential quantum error correction – below threshold!

Errors are one of the biggest challenges in quantum computing because qubits, the computing units in quantum computers, tend to quickly exchange information with their surroundings, making it difficult to protect the information needed to perform a calculation. Typically, the more qubits you use, the more errors occur and the system becomes classic.

Today we published results in Nature that prove this The more qubits we use in Willow, the more reduce Mistakeand the more quantum the system becomes. We tested increasingly larger arrays of physical qubits, scaling from a grid of 3×3 encoded qubits, to a 5×5 grid, to a 7×7 grid – and each time, using our latest advances in quantum error correction, we were able to reduce the error rate in half. In other words: we have achieved an exponential reduction in the error rate. This historic achievement is what experts call “below threshold” – the ability to reduce errors while increasing the number of qubits. You must demonstrate that you are below the threshold to show real progress in correcting errors. This has been a prominent challenge since Peter Shor introduced quantum error correction in 1995.

Other scientific “firsts” are also involved in this result. For example, it is also one of the first convincing examples of real-time error correction in a superconducting quantum system – crucial to any useful calculation, because if you can’t correct errors fast enough, they ruin your calculation before it’s finished. And it’s a “beyond break-even” demonstration, where our qubit arrays last longer than the individual physical qubits, a sure sign that error correction improves the system overall.

As the first subthreshold system, this is the most compelling prototype for a scalable logic qubit to date. It’s a strong sign that useful, very large quantum computers can actually be built. Willow brings us closer to running practical, commercially relevant algorithms that cannot be replicated on traditional computers.

10 septillion years on one of the fastest supercomputers today

As a measure of Willow’s performance, we used the RCS (Random Circuit Sampling) benchmark. RCS was developed by our team and is now widely used as a standard in this area. It is the classically hardest benchmark that can be performed on a quantum computer today. You can think of this as an entry point for quantum computing – testing whether a quantum computer can do something that wouldn’t be possible on a classical computer. Any team building a quantum computer should first check whether it can beat classical computers on RCS. Otherwise, there is definitely reason to be skeptical as to whether it can be used to solve more complex quantum tasks. We have consistently used this benchmark to assess progress from one generation of chips to the next – we reported Sycamore results in October 2019 and recently again in October 2024.

Willow’s performance on this benchmark is astonishing: it performed a calculation in under five minutes that would take one of today’s fastest supercomputers 10 minutes²⁵ or 10 septillion years. If you want to write it down, it’s 10,000,000,000,000,000,000,000,000 years. This staggering number exceeds the time scales known in physics and far exceeds the age of the universe. It lends credence to the idea that quantum computation occurs in many parallel universes, consistent with the idea that we live in a multiverse, a prediction first made by David Deutsch.

These latest results for Willow, as shown in the chart below, are our best to date, but we will continue to make progress.