Google Quantum Chip Wins Magic Square Game at 98.3%
Willow's four-qubit result tops the 88.9% classical ceiling in a study published in Science.
Updated on Nov. 19, 2026 in Quantum Computing

Google Quantum AI's 105-qubit Willow processor won the four-qubit Mermin-Peres magic square game at a 98.3% success rate, above the 88.9% classical strategy ceiling. The peer-reviewed study was published in Science on November 19, 2026, and featured on its cover.
Why it matters
The result provides experimental evidence that quantum hardware can exceed classical limits on a contextuality test, a step toward demonstrating practical quantum advantage. Because the team also scaled the tests across all 105 qubits, rival labs will likely be measured against this demonstration.
Willow is a 105-qubit superconducting transmon processor, and the team ran nine game assignments, each executed 100,000 times, achieving a 98.3% success rate against a classical ceiling of 88.9%. Residual hardware noise accounts for most of the losses, and the study does not report performance beyond the tested assignments.
The players
Google Quantum AI
Google Quantum AI is the quantum computing research division of Google, which developed the Willow superconducting processor and has pursued error-corrected quantum hardware for several years.
The details
The study extended contextuality tests to multi-player GHZ-type games and hidden linear function problems, and scaled the contextuality tests to all 105 qubits on the chip. The hidden linear function work relied on a shallow-circuit algorithm. A preprint version appeared on arXiv in December 2025 before peer-reviewed publication.
Timeline
December 2024: Google Quantum AI announced the Willow chip.
December 2025: A preprint version of the study appeared on arXiv.
November 19, 2026: Science published the peer-reviewed study, featuring Willow on its cover.
Quantum Race
Google's 2019 Sycamore supremacy experiment set an early benchmark for superconducting quantum hardware, and this Willow result moves the comparison toward a verifiable test that classical methods cannot match. It positions Google's hardware program as a direct successor to that earlier demonstration, with rival quantum labs now measured against the scale of the Willow work.
The result is a research milestone rather than a consumer product, so readers will not see immediate changes to devices or services. Its practical relevance depends on whether future quantum systems can extend contextuality advantages to tasks that matter outside the laboratory.
The takeaway
Verifiable quantum advantage tests like the magic square game give researchers a way to separate hardware progress from classical shortcuts. Watch for independent replications and larger-scale follow-ups to judge whether the result generalizes.
Further reading
For broader context on hardware and error correction, see Quantum Computing coverage on Wisevoter.



