Xanadu

Photonic Founded 2016 Toronto, ON, Canada

Overview

Photonic quantum computing using squeezed light states. Room-temperature operation with focus on continuous-variable quantum computing and Gaussian boson sampling.

Current System: 216 qubits
Funding: SPAC merger with Crane Harbour at ~$3.6B enterprise value — first pure-play photonic quantum IPO.
Website: https://www.xanadu.ai

Key Milestones

  • 2016: Xanadu founded by Christian Weedbrook
  • 2020: PennyLane quantum ML framework released (open source)
  • 2021: Borealis processor achieves quantum computational advantage
  • 2023: 216-mode X8 processor demonstrated
  • 2024: Announced path to public listing

Technology: Photonic Quantum Computing

Xanadu uses photons as qubits, manipulated via optical components (beam splitters, phase shifters, squeezers). Key advantages over superconducting/ion trap systems:

  • Room temperature operation (no cryogenics)
  • Network-native (photons travel through fiber)
  • Low error rates for certain operations (linear optics)

Challenge: Photonic gates require measurement and feedforward, making universal quantum computing difficult. Xanadu’s solution: continuous-variable (CV) quantum computing and Gaussian boson sampling (GBS).

Borealis & X8 Processors

Borealis (2021): 216-mode photonic processor that demonstrated quantum computational advantage via Gaussian boson sampling. Sampled from distributions classical supercomputers cannot efficiently compute.

X8 (2023): Next-generation photonic architecture with improved squeezing, lower loss, and programmable interferometers.

Roadmap: Fault-tolerant photonic quantum computing via measurement-based quantum computation (MBQC) using cluster states.

PennyLane Ecosystem

Xanadu’s open-source quantum machine learning framework. Integrates with PyTorch, TensorFlow, JAX for hybrid quantum-classical ML.

Adoption: Used by researchers at Google, IBM, universities. Positions Xanadu as software-first company, not just hardware vendor.

Path to Public Listing

Xanadu announced plans for public listing in 2024, positioning as first photonic quantum company to go public. Revenue model: cloud access, enterprise licenses, government contracts.

Competitive Position

vs. Superconducting (IBM, Google):
Room temperature, network-native, but limited gate set. Photonic systems excel at specific tasks (sampling, quantum communication) but lag in universal computation.

vs. PsiQuantum:
Both photonic, but different approaches. Xanadu: near-term systems available now. PsiQuantum: fault-tolerant system, not available until late 2020s.

Applications

  • Gaussian boson sampling (computational advantage demonstrations)
  • Quantum chemistry (vibronic spectra, molecular simulation)
  • Quantum machine learning (variational circuits, kernel methods)
  • Quantum networking (photons naturally suited for fiber links)

Target industries: Pharmaceuticals (molecular simulation), finance (optimization), aerospace (engineering simulations).