Quantum network modelling, in preview
QNetwork is an emulated preview. It lists quantum backends with reference data and compares trapped-ion configurations for a circuit with classical models. It does not dispatch jobs, create physical entanglement or choose a backend for you.
What runs today
These routes answer on the QCOS API. Each one returns a model output or reference data. None of them runs a circuit.
Backend listing
Reference entries for superconducting, trapped-ion and simulator backends, with technology and connectivity. The response is flagged demo_mode: the figures are reference data, not live readings.
Trapped-ion configuration sweep
Compares ion species, vibrational modes and qubit encodings for a circuit of a given size, using a classical model and one fixed scoring policy. The response is labelled EMULATED. No circuit runs, and the output is a model result, not a validated backend choice.
Ion species reference
Reference properties for trapped-ion species: coherence times, gate fidelities and gate durations.
Model topology
A three-node model network. The response is labelled EMULATED and states that no physical quantum resources are involved.
Not offered yet
- •Automatic backend selection
- •Load balancing across backends
- •Automatic failover
- •Running one circuit on several backends and aggregating the results
- •Entanglement, teleportation and remote gates between machines
Automatic backend selection comes later, and in order: observability first, then measurements, then a corpus of validation records that any selector has to be checked against.
To run a circuit, use QCOS: you name the backend for every job.
Backend technologies in the listing
QNetwork lists these backend technologies. It does not submit jobs to them.
Superconducting
ModalityTrapped ion
ModalitySimulator
ModalityQuick Start
Read the listing and run a sweep with the Python SDK. Nothing in this sample runs a circuit.
# pip install softqcos-sdk
from softqcos_sdk import QCOSClient
client = QCOSClient(api_key="your-api-key")
# The listing: reference data, flagged as such
listing = client.network.backends()
print(listing["demo_mode"]) # True
for backend in listing["backends"]:
print(backend["name"], backend["provider"], backend["technology"])
# A classical model of trapped-ion configurations for a small circuit
sweep = client.network.phase0_sweep(
num_qubits=4,
single_qubit_gates=4,
two_qubit_gates=3,
circuit_depth=4,
)
print(sweep["mode"]) # "EMULATED"
print(sweep["total_configurations"], sweep["valid_configurations"])Tell us what you need from a quantum network
QNetwork is a preview. Tell us the workload you have in mind and we will tell you what runs today and what does not.