Put your QPU online
Software to automatically calibrate and control quantum processors
The problem
Today, engineers bring quantum devices online by hand. They tune every qubit, one parameter at a time.
It takes weeks. It is expensive. It does not scale.
As processors grow in size and complexity, this approach breaks down. The industry needs software that connects low-level hardware control to high-level workflows.
How Control works
Calibration is a series of experiments: measure, analyse, decide, repeat. Control automates this workflow with ML-powered analysis.
Connect your hardware
Wire in your control electronics and define your device in a single configuration file. Map instruments, channels, and qubit parameters. Control takes it from here.
Perform measurement
Control drives your instruments to run characterisation experiments, e.g. spectroscopy sweeps, Rabi oscillations, resonator readout, collecting raw data from the quantum device.
Analyse with ML models
ML models trained on quantum measurement data interpret each result automatically. They identify noisy features and replace hours of manual analysis.
Decide what’s next
Based on the analysis, Control decides the next action. Adjust a parameter and re-measure, compensate for crosstalk, or move to the next calibration step. The loop repeats until every parameter is optimised.
Experiment complete
The calibration experiment finishes and results are stored. Control moves on to the next experiment in the sequence, repeating the loop until the full calibration routine is done.
Better with experience
Each calibration pass makes the next one more reliable and accurate through continuous refinement.
Optimize uptime & performance
Detect drift early, re-calibrate automatically, and tune gate fidelities with closed-loop feedback.
Scales to billions of qubits
Parallelized architecture with no bottlenecks, same software from tens to billions of qubits.
# Models SDK - ML Models from conductorquantum import ConductorQuantum # Initialize client client = ConductorQuantum(token=TOKEN) # Load Charge Stability Diagram data data = np.load("data.npy") # shape (n, m) # Detect the transition lines result = client.models.execute( model="charge-stability-diagram-transition-detector-v3", data=data ) # Access the transition lines result transition_lines = result.output["transition_lines"]
Keep your routines,
automate the analysis
Our Models SDK plugs into your existing workflow. Send measurement data and get structured results back.
Models API docs