Overview
Public access to pulse-level and control-electronics interfaces in commercial quantum computing has become bifurcated, and in my opinion will only continue to do so. The largest superconducting cloud platform, IBM Quantum, removed pulse-level control from all production QPUs in February 2025. By far the most important challenge of the coming years will be not in novel physics advancements for further manufacturing techniques, but rather firm software engineering infrastructure to democratize quantum computing.
Mid-tier superconducting vendors and the more open neutral-atom platforms have moved in the opposite direction. This paper documents the bifurcation row by row.
The paper surveys thirteen commercial vendors across superconducting, trapped-ion, neutral-atom, and photonic modalities, grading each on six axes of openness at the control plane. That being a definition I stole from cloud computing. It is the layer between gate-level circuit specifications and physical control electronics.
The catalog ships as a separate machine-readable artifact under CC-BY-4.0, intended to be cited, forked, and updated as vendor access policies shift.
Vendors include Atom Computing, Google, IBM, IonQ, IQM, OQC, ORCA, Pasqal, PsiQuantum, Quantinuum, QuEra, Rigetti, and Xanadu.
What the paper does and does not do
The paper does NOT propose an architecture or a reference implementation. It describes that the field has lost as the access landscape has shifted over the years, and what (admittedly self described) open access at this layer would have to look like.
I designed the grading rubric to be intentionally narrow: each axis is graded only on public documentation, with per-cell URLs and accessed-on-dates so that the catalog can be re-graded as vendor policies change.
Additional Thoughts
This paper is a bit naive in what is in the best intersts of vendors to provide/not provide. I understand it is certainly in the best interst of major quantum computing vendors to have parts of their stacks closed source, that is just business. As with Linux however, some parts are too important to keep closed source, and quite frankly most companies would not want to be solely responsible for such a layer. I have another project in the works for a kernel type layer that sits at the control stack level, that then facilitates these propritary vendor pulse stacks.
Links
Citation
Malarchick, R. (2026). "Measuring Control-Plane Openness in Near-Term Quantum Computing: A Rubric, Its Validation, and an Application to Thirteen Vendor Stacks." arXiv:2605.15233.