Experiment with Qiskit Runtime¶

Overview¶

Qiskit Runtime speeds up processing time by combining classical and quantum computing in a streamlined architecture.

In this first version you can explore running simple circuits or more complex variational algorithms (based on VQE). Additionally, you can test it using our quantum simulators on the cloud, so that you can quickly get up to speed with this new way to run Qiskit programs. With Qiskit Runtime, you can keep your quantum program on the cloud and run it any time.

View available Qiskit Runtime programs¶

On the Services page, click the Programs tab to view available programs. Click any program to learn more about it, and optionally choose a backend and provider to generate a custom program template and open it directly in Quantum Lab. You can then modify it as necessary and use it to execute circuits.

Run a Qiskit Runtime program¶

In this example we will use a preloaded Qiskit Runtime program called sampler. We will first build a circuit using Qiskit, then execute it with Qiskit Runtime, taking advantage of its low-latency architecture.

The sampler program is one of the primitive programs (another is the estimator). These primitives provide methods that make it easier to build modular algorithms and other higher-order programs. They provide a seamless way to leverage the latest optimizations in IBM Quantum hardware and software.

The Sampler primitive lets you more accurately contextualize counts. It takes a user circuit as an input and generates a readout of quasi-probabilities. This enables you to more efficiently evaluate the possibility of multiple relevant data points in the context of destructive interference.

qiskit-ibm-runtime now provides a simple Sampler class which allows access to the Sampler primitive service.

Instantiate the QiskitRuntimeService from one of the saved accounts like below.

from qiskit_ibm_runtime import QiskitRuntimeService, Sampler

# Save your credentials on disk.
# QiskitRuntimeService.save_account(channel='ibm_quantum', token=<IBM Quantum API key>)

service = QiskitRuntimeService()

A Qiskit Runtime session allows you to cache data, such as parameterized circuits, on the server side for the duration of a session. You can then pass in only parameters during subsequent calls.

Let’s prepare the quantum circuits as shown below.

from qiskit import QuantumCircuit

bell = QuantumCircuit(2)
bell.h(0)
bell.cx(0, 1)
bell.measure_all()

print(bell.draw())

        ┌───┐      ░ ┌─┐
   q_0: ┤ H ├──■───░─┤M├───
        └───┘┌─┴─┐ ░ └╥┘┌─┐
   q_1: ─────┤ X ├─░──╫─┤M├
             └───┘ ░  ║ └╥┘
meas: 2/══════════════╩══╩═
                      0  1

The quantum circuits can then be passed as lists to the Sampler class, which instantiates and opens a session with the Sampler primitive service.

sampler can then be called multiple times to write different circuit_indices (a list of circuit indices) and parameter_values (an optional list of concrete parameters to be bound) to the session and read the calculated quasi-probabilities. The result is an instance of SamplerResult class in qiskit-terra and it contains quasi-probabilities and metadata.

with Sampler(circuits=[bell]*3, service=service, options={ "backend": "ibmq_qasm_simulator" }) as sampler:
    result1 = sampler(circuit_indices=[0, 1, 2], shots=2000)
    print(result1)

    result2 = sampler(circuit_indices=[0, 1, 2], shots=3000)
    print(result2)

    result3 = sampler(circuit_indices=[0, 1, 2], shots=4000)
    print(result3)

SamplerResult(quasi_dists=[{'00': 0.5155, '11': 0.4845}, {'11': 0.5175, '00': 0.4825}, {'00': 0.4915, '11': 0.5085}], metadata=[{'header_metadata': None, 'shots': 2000}, {'header_metadata': None, 'shots': 2000}, {'header_metadata': None, 'shots': 2000}])
SamplerResult(quasi_dists=[{'11': 0.5043333333333333, '00': 0.49566666666666664}, {'00': 0.501, '11': 0.499}, {'00': 0.493, '11': 0.507}], metadata=[{'header_metadata': None, 'shots': 3000}, {'header_metadata': None, 'shots': 3000}, {'header_metadata': None, 'shots': 3000}])
SamplerResult(quasi_dists=[{'00': 0.495, '11': 0.505}, {'11': 0.50075, '00': 0.49925}, {'11': 0.49825, '00': 0.50175}], metadata=[{'header_metadata': None, 'shots': 4000}, {'header_metadata': None, 'shots': 4000}, {'header_metadata': None, 'shots': 4000}])

View your Qiskit program jobs¶

You can view details about running and completed jobs by opening the Jobs page from the application switcher ( ), then clicking the Program jobs tab.

Qiskit Runtime tutorials and other resources¶

To learn more, visit the Qiskit Runtime Github repository. You can also implement Qiskit Runtime by using the public REST API.

Explore these Qiskit Runtime tutorials:

• Get started with the Sampler primitive

• Get started with the Estimator primitive

• Database search with Grover’s algorithm using the Sampler primitive

• Submitting user-transpiled circuits using primitives

• Variational Quantum Eigensolver with Estimator primitive

• Spectroscopic Eigensolver Algorithm with Sampler

• Qiskit Runtime Introduction (Open directly in IBM Quantum Lab here)

• How to use Qiskit Runtime VQE for chemistry simulation (Open directly in IBM Quantum Lab here)

• How to use Qiskit Runtime Quantum Kernel Alignment (QKA) for Machine Learning (Open directly in IBM Quantum Lab here)

Limitations¶

Qiskit Runtime glossary¶