Insider Brief
- Xanadu has outlined an updated technology roadmap targeting fault-tolerant quantum computing by 2028–2029 and more than 1,000 logical qubits by 2031.
- The roadmap targets reducing the company’s photon-loss indicator from 24.1x in 2026 to 1.0x by 2030 while improving logical error rates to 10⁻¹⁶.
- Xanadu also plans to build a quantum data center in 2029–2030 and expand its PennyLane software platform across quantum hardware and simulation environments.
PRESS RELEASE – Xanadu Quantum Technologies Limited (“Xanadu” or the “Company”) (NASDAQ: XNDU) (TSX: XNDU), a leading photonic quantum computing company, will host its Technology Roadmap Presentation today at 1:00 p.m. ET, featuring presentations from members of its executive and technical leadership team on the Company’s technology roadmap and path to commercial utility-scale quantum computing.
A live webcast and accompanying presentation materials will be available through the Company’s investor relations website at investors.xanadu.ai, with a replay available following the event.
In connection with the event, Xanadu is providing an expanded, milestone driven technology roadmap reflecting progress across its photonic quantum computing architecture and PennyLane software platform.
Hardware: Advancing Toward Fault-Tolerant Quantum Computing
Xanadu’s hardware roadmap is focused on reducing loss, the primary source of error in its photonic architecture, to a level where quantum error correction can increasingly correct logical errors. The Company is working toward this through improvements in materials, fabrication, packaging and system architecture, alongside continued advances in quantum error correction and architectural design.
Reducing loss: Xanadu is targeting continued improvements in photon loss, a key measure of errors, through advances in component performance, manufacturing processes and system design. The roadmap targets a reduction in the Company’s key loss technical indicator from 24.1x by 2026 to 1.0x by 2030. These improvements, if achieved, are expected to progressively improve logical qubit quality and enable quantum error correction to more effectively correct errors as the system scales.
Advancing error correction: Xanadu’s fault-tolerant architecture is built around a concatenated error-correction approach that combines GKP encoding with quantum low-density parity-check (qLDPC) codes. As physical photon loss improves and falls below the required error-correction threshold, the architecture is designed to increasingly correct logical errors, with logical error rates targeted to improve to 10⁻16 by 2030. This improvement in logical qubit quality is expected to provide the foundation for scaling the number of usable logical qubits, with Xanadu targeting up to 200 logical qubits by 2029, up to 500 by 2030, and 1,000+ by 2031.
Scaling toward fault tolerance: Xanadu’s updated roadmap includes the following projected technical milestones and projects:
- Qubit Factory build: 2026–2027
- Fault tolerance milestone: 2028-2029
- Up to 200 logical qubits by 2029, scaling to up to 500 by 2030 and 1,000+ by 2031
- Logical error rate to improve from roughly 10⁻3-10⁻8 in 2028 to 10⁻16 by 2030
- Quantum data center: 2029–2030
Xanadu’s photonic architecture is designed to operate at room temperature, be manufactured using silicon processes and scale through modular, networked systems compatible with existing telecom infrastructure. The Company is also building out its 158,000-square-foot Toronto manufacturing facility, Inception, to support testing, heterogeneous integration, photonic integrated circuit packaging and rack-level module assembly.
“We’ve cut our loss ratio dramatically, and that progress is what gives us confidence in reaching fault tolerance in 2028-2029,” said Christian Weedbrook, Founder and CEO of Xanadu. “What’s left is largely a materials, fabrication and systems engineering challenge, along with further architecture optimizations, areas we believe our approach is built to handle as we scale from fault tolerance to a networked quantum data center by the end of the decade.”
Software: PennyLane Extends Its Position as a Leading Quantum Development Platform
Alongside its hardware roadmap, Xanadu continues to expand PennyLane, its open-source platform for building quantum applications in Python and running them across simulators and quantum processors from multiple hardware vendors.
Every major shift in computing has been shaped not only by advances in hardware, but by the software platforms that enable developers to build on them. Xanadu is building PennyLane with the goal of becoming that foundational software layer for quantum computing.
We believe quantum application development still runs into real constraints: quantum programming expertise remains scarce, early tooling often locks developers into a single hardware vendor’s stack, and many programming languages cap what the underlying machine can do rather than exposing its full power. Documentation is another barrier: a 2025 industry survey found poor documentation was developers’ leading reason for abandoning a tool, ahead of price.
PennyLane is built to remove those constraints. It’s easy to use, the Catalyst compiler translates hybrid quantum-classical programs down to the target device, and the Lightning simulator suite runs the workload on a CPU, a GPU, an HPC cluster, or a quantum processor, including Xanadu‘s own hardware. The result is a single framework developers can build in once and deploy across the industry’s major hardware modalities, including superconducting, trapped ion, neutral atom and photonic systems, and adoption has followed:
- PennyLane was used by 30.8% of developers surveyed over the past year, the second-highest usage share among quantum development platforms
- More than 1,890 public code repositories declare PennyLane as a dependency
- The platform family sees approximately 1.1 million package installs per month
- 51 simulator and hardware devices are reachable from a single PennyLane API
- More than 150 university partners across more than 35 countries use PennyLane in research and teaching
We believe PennyLane’s ease of use and first-class developer experience are making it a platform of choice for researchers, developers, and enterprises building quantum capabilities. Organizations are using it to train teams, build quantum expertise, and explore real-world applications, including the Quantum Talent Pipeline partnership with Lockheed Martin, as well as applied research collaborations with organizations including Rolls-Royce, Toyota, Volkswagen, AMD and Mitsubishi Chemical Group.
“Every developer, researcher, and enterprise building with PennyLane today is a potential customer when our hardware comes online. That’s the real value of PennyLane,” said Rafal Janik, Chief Operating Officer of Xanadu. “It gives us a foothold in the quantum ecosystem today, allowing us to build relationships that have the potential to evolve into applied quantum solutions and, ultimately, Xanadu hardware and cloud services.”
Commercial model: Xanadu sees multiple potential paths to monetizing its software ecosystem over time, including enterprise training and enablement, applied algorithm development, future cloud-delivered services and hardware sales to customers already using its software tools. These paths feed into a broader commercialization strategy as the Company targets meaningful end-customer commercialization by 2029–2030, underpinned by a full-stack platform spanning proprietary hardware and software and strategic partnerships.
Financial Position and Government Funding
Capitalization: We believe Xanadu is well-capitalized to execute on this roadmap, with approximately US$686 million of total available funding: US$312.8 million of cash and cash equivalents and US$232.8 million of remaining capacity under the synthetic at-the-market equity facility (SATM) with Yorkville Advisors, each as of June 30, 2026, plus approximately US$140.2million (CAD $195 million) of confirmed Government of Canada funding under Project OPTIMISM.
