The Remote device

The 'qiskit.remote' device is a generic adapter to use any Qiskit backend as interface for a PennyLane device.

To access IBM backends, we recommend using Qiskit Runtime.

from qiskit_ibm_runtime import QiskitRuntimeService

QiskitRuntimeService.save_account(channel="ibm_quantum", token="<YOUR_IBMQ_TOKEN>")

# To access saved credentials for the IBM quantum channel and select an instance
service = QiskitRuntimeService(channel="ibm_quantum", instance="my_hub/my_group/my_project")
backend = service.least_busy(operational=True, simulator=False, min_num_qubits=<num_qubits>)

dev = qml.device('qiskit.remote', wires=<num_qubits>, backend=backend)

Note

Certain third-party backends may be using the deprecated Provider interface, in which case you can get the backend instance from providers with complex search options using their get_backend() method. For example:

import pennylane as qml

def configured_backend():
    backend = SomeProvider.get_backend(...)
backend.options.update_options(...) # Set backend options this way
return backend

dev = qml.device('qiskit.remote', wires=2, backend=configured_backend())

After installing the plugin, this device can be used just like any other PennyLane device for defining and evaluating QNodes. For example, a simple quantum function that returns the expectation value of a measurement and depends on three classical input parameters can be decorated with qml.qnode as usual to construct a QNode:

@qml.qnode(dev)
def circuit(x, y, z):
    qml.RZ(z, wires=[0])
    qml.RY(y, wires=[0])
    qml.RX(x, wires=[0])
    qml.CNOT(wires=[0, 1])
    return qml.expval(qml.PauliZ(wires=1))

You can then execute the above quantum circuit to get the expectation value of a Pauli operator.

circuit(0.2, 0.1, 0.3)

The 'qiskit.remote' device also supports the use of local simulators such as FakeManila.

from qiskit_ibm_runtime.fake_provider import FakeManilaV2
backend = FakeManilaV2()

# You could use an Aer simulator instead by using the following code:
# from qiskit_aer import AerSimulator
# backend = AerSimulator()

dev = qml.device('qiskit.remote', wires=5, backend=backend)

Device options

The 'qiskit.remote' device uses the EstimatorV2 and the SamplerV2 runtime primitives to execute the measurements. To set options for transpilation or runtime, simply pass the keyword arguments into the device.

dev = qml.device(
    "qiskit.remote",
    wires=5,
    backend=backend,
    resilience_level=1,
    optimization_level=1,
    seed_transpiler=42
)
# to change options, re-initialize the device
dev = qml.device(
    "qiskit.remote",
    wires=5,
    backend=backend,
    resilience_level=1,
    optimization_level=2,
    seed_transpiler=24
)

This device is not compatible with analytic mode, so an error will be raised if shots=0 or shots=None. The default value of the shots argument is 1024. You can set the number of shots on device initialization using the shots keyword, or you can choose the number of shots on circuit execution.

dev = qml.device("qiskit.remote", wires=5, backend=backend, shots=4096)

@qml.qnode(dev)
def circuit(x, y, z):
    qml.RZ(z, wires=[0])
    qml.RY(y, wires=[0])
    qml.RX(x, wires=[0])
    qml.CNOT(wires=[0, 1])
    return qml.expval(qml.PauliZ(wires=1))

# Runs with 4096 shots
circuit(0.2, 0.1, 0.3)

# Runs with 10000 shots
circuit(0.2, 0.1, 0.3, shots=10000)

devices/remote

 

Download Python script

 

Download Notebook

 

View on GitHub