A quantum simulation framework that uses IBM's superconducting processors to test how networks of coupled oscillators synchronise. Classical Kuramoto theory in, noisy hardware out, and the gap in between is where the physics lives.
On ibm_kingston (Heron r2, 156 qubits), a 342-circuit campaign across 8 Trotter depths shows that the odd parity sector of the XY Hamiltonian's dynamical Lie algebra is systematically more robust to decoherence than the even sector. Peak asymmetry is +17.5 % at depth 6, mean is $(10.8 \pm 1.1)\,\%$ for depths $\ge 4$, Fisher's combined p-value across all eight depths is below numerical precision ($p \ll 10^{-16}$), and the observed magnitude at large depth falls inside the 4.5–9.6 % range predicted in advance by our classical Rust simulator.
This is the first direct empirical confirmation that the $\mathrm{DLA}(H_{XY}) = \mathfrak{su}(2^{n-1}) \oplus \mathfrak{su}(2^{n-1})$ block structure is distinguishable in hardware, and it hands us the exact noise profile needed to calibrate symmetry-guided error mitigation (GUESS) for the Phase 2 scaling campaign.
Seven open-access preprints (25 May 2026) document the hardware campaigns, methods, and software behind this project. Each is archived on Zenodo with a citable DOI and carries the claim boundaries stated in its own abstract. These are preprints under the project's evidence discipline — not peer-reviewed journal articles.
ibm_kingston. Replication on a second Heron r2 device (ibm_marrakesh) is planned in the Phase 2 campaign to rule out any residual device-specific effect.scpn-quantum-control source is accessible to commercial licence holders, academic researchers, contributors under CLA, and peer reviewers. Request access via protoscience@anulum.li. The PyPI package itself is public.