Blog
Practical insights on quantum device selection, benchmarking, and the WCPP framework.
Why we published a negative result on ZNE before publishing the positive one.
Our calibration-informed zero-noise extrapolation failed four different ways. We reported it as Appendix D.1 of the v1.2 paper. Here's the reasoning — and what a failed experiment taught us about the forward equation.
Two of our four capability axes were perfectly rank-correlated on 9 devices. Here's what we did.
A reviewer asked whether WCPP's four axes are empirically independent. ρ(Γ, F) turned out to be −0.56 — good. ρ(Γ, Φ) turned out to be Spearman +1.00 (p ≈ 5.5×10⁻⁶) — not good. How we responded, and why this is in a blog post.
We ran 100 circuits across 4 quantum vendors. Here's the r(τ) you can't unsee.
Pearson r between Qlro's predicted and observed fidelity drops from 0.96 to 0.69 over three days of calibration age on superconducting hardware. A first quantitative point on a decay curve nobody has characterised yet.
Quantum Device Selection in 2026: A Practical Guide
How to choose the right quantum computer for your workload — with real benchmark data across IBM, IQM, Rigetti, and IonQ.