Photonic Publishes SHYPS QLDPC Logic Research in Nature Communications

Key Takeaways

Peer-Reviewed Publication: Nature Communications has published Photonic’s paper describing SHYPS codes for efficient QLDPC logic and error correction.

Resource Overhead: In simulated comparisons, SHYPS used 2× to 3.5× fewer physical qubits than comparably scaled rotated surface codes.

Architecture Constraint: The SHYPS approach requires connectivity beyond nearest-neighbor architectures, making it particularly suited to high-connectivity systems such as Photonic’s Entanglement First architecture.

On August 25, 2026, Photonic Inc. announced that Nature Communications has published “Computing Efficiently in QLDPC Codes,” a paper on Subsystem Hypergraph Product Simplex (SHYPS) codes. The work, first posted as a preprint in February 2025, reports that this family of quantum low-density parity check (QLDPC) codes can implement logical Clifford operations and error correction with fewer physical qubits than comparable surface-code encodings in the simulated comparisons. Authors are based at Photonic, with coauthors affiliated with the University of Edinburgh and Freie Universität Berlin. The journal lists publication on 20 May 2026 and the version of record on 27 July 2026 (DOI: 10.1038/s41467-026-73061-9).

SHYPS Code Construction and Performance

QLDPC codes encode multiple logical qubits in a single block with sparse parity checks, a structure long expected to cut the physical-qubit overhead of quantum error correction relative to planar surface codes. Prior QLDPC work emphasized memory. The Photonic paper targets logical computation. SHYPS codes are formed by combining the subsystem hypergraph product construction with classical simplex codes. Parameterized by an integer r ≥ 3, each instance has parameters [n, k, d] = [(2r − 1)2, r2, 2r−1] and weight-3 gauge generators. Simulated instances include:

[49, 9, 4] (r = 3), compared with a scaled rotated surface code [[81, 9, 3]]

[225, 16, 8] (r = 4), compared with [[784, 16, 7]]

[961, 25, 16] (r = 5), evaluated in additional memory simulations without the same surface-code comparison

In the simulated comparisons, the [49, 9, 4] SHYPS code used about 2× fewer physical qubits than [[81, 9, 3]], while [225, 16, 8] used about 3.5× fewer than [[784, 16, 7]]. Circuit-level simulations of depth-126 logical circuits on two [49, 9, 4] blocks show near-memory logical performance. The construction implements the full Clifford group with transversal and fold-transversal operations so that any m-qubit Clifford can be executed in at most O(m) syndrome-extraction rounds, with depth independent of code distance. Combined with known methods for implementing T gates by state injection, the authors present this as a route to universal quantum computation using SHYPS. Simulation circuits and data are released on GitHub and Zenodo. Lead author Alexander J. Malcolm and coauthors include Joschka Roffe of Edinburgh and Photonic and Armanda O. Quintavalle of Freie Universität Berlin.

Connectivity Requirements and Commercial Path

Photonic states that the SHYPS overhead reductions reported in simulation are available only to hardware with connectivity beyond nearest-neighbor layouts. The company’s Entanglement First architecture is built around silicon T-centre qubits with optical interconnects, including on-chip and telecom-fiber links intended to support non-local parity checks. Dr. Stephanie Simmons, Chief Quantum Officer at Photonic, said efficient QLDPC logic is “no longer a theoretical promise, it’s a demonstrated result, with real implications for architectures and timelines.” Photonic describes SHYPS as the first QLDPC code family demonstrated to efficiently perform both quantum logic and error correction. She added that the company has continued work on QLDPC codes since the preprint. Photonic’s 2025 launch materials claimed reductions of up to 20× at larger projected scales; the peer-reviewed paper reports 2× and 3.5× reductions in its simulated comparisons.

The result is relevant to system-level resource estimates because surface-code lattice surgery typically incurs depth that grows with distance. The SHYPS framework provides an O(1) Clifford compiling ratio, with logical Clifford depth independent of code distance. That distinction matters for architectures that can supply the required connectivity without collapsing qubit density or cycle time. Photonic, headquartered in Vancouver, British Columbia, with operations in the United States and the United Kingdom, positions SHYPS as part of a distributed computing stack aimed at materials, chemistry, and security workloads. Additional technical notes are collected on the company’s error-correction page and in the original SHYPS announcement. Independent wire coverage of the journal publication appeared via GlobeNewswire and HPCwire.

Bottom Line

Peer-reviewed simulations show SHYPS can implement logical Clifford operations with fewer physical qubits than the compared surface-code encodings, assuming the connectivity required by the architecture.

Find out more here.

Further articles, reports, and the latest quantum computing news may be found at The Qubit Report.

Related Articles

Get The Qubit Report delivered straight to your inbox.

Sign up to receive our newsletter and other reports.

We keep your data private and share your data only with third parties that make this service possible. Read our privacy policy for more info.