3

It is claimed that the fake backends of IBM will have noise similar to the real backends. However, to what extent can we expect them to produce the same results?

I have run my quantum circuit on the real backend, ibm_kyoto and the result differs vastly from the fake backends (FakeCairoV2, FakeTorino) which I test my algorithm with.

Do the real backends have much more noise than the fake ones?

FDGod
  • 3,000
  • 2
  • 7
  • 31
Deren Liu
  • 243
  • 1
  • 4

1 Answers1

1

Fake backends from qiskit-IBM-runtime package contains calibration results from a snapshot that is hardcoded in the source code. To get a more realistic simulation results, you can use qiskit-aer with the latest calibration results:

from qiskit_aer import AerSimulator
from qiskit_ibm_runtime import QiskitRuntimeService

service = QiskitRuntimeService()
real_backend = service.backend('ibm_kyoto')
aer_simulator = AerSimulator.from_backend(real_backend)

You can read more in the documentation about different local testing modes: https://docs.quantum.ibm.com/verify/local-testing-mode

Junye Huang
  • 712
  • 8
  • 18