Benchmarks Tutorial

This tutorial demonstrates how to run quantum benchmarks to evaluate device performance.

Randomized Benchmarking

Estimate average gate error rate:

from microquantum import RandomizedBenchmarking

rb = RandomizedBenchmarking(num_qubits=1, seed=42)
result = rb.run(sequence_lengths=[1, 2, 4, 8, 16], num_samples=20)

print(f"Average gate fidelity: {result.average_gate_fidelity:.4f}")
print(f"Error per gate: {result.error_per_gate:.4f}")

Cross-Entropy Benchmarking

Measure quantum advantage potential:

from microquantum import CrossEntropyBenchmarking

xeb = CrossEntropyBenchmarking(
    num_qubits=4,
    depths=[5, 10, 20],
    num_circuits=5,
    seed=42,
)
result = xeb.run()

for depth, fid in zip(result.depths, result.fidelities):
    print(f"Depth {depth}: XEB fidelity = {fid:.4f}")

Quantum Volume

Measure effective quantum volume:

from microquantum import QuantumVolumeBenchmark

qv = QuantumVolumeBenchmark(max_qubits=5, seed=42)
result = qv.run()
print(f"Quantum volume: {result.value}")

Gate Set Tomography

Characterize individual gate fidelity:

from microquantum import GateSetTomography
from microquantum.core import Operator

gst = GateSetTomography([Operator.H(), Operator.X()])
result = gst.run()

for name, fid in zip(result.gate_names, result.gate_fidelities):
    print(f"{name}: fidelity = {fid:.4f}")

CLOPS

Measure circuit execution speed:

from microquantum import CLOPSBenchmark

clops = CLOPSBenchmark(num_qubits=3, num_layers=5, num_circuits=20, seed=0)
result = clops.run()
print(f"CLOPS: {result.value:.0f}")