diff --git a/docs/manuals/characterization/stark_experiment.rst b/docs/manuals/characterization/stark_experiment.rst
index 23b6c56d4e..8233a6ac6f 100644
--- a/docs/manuals/characterization/stark_experiment.rst
+++ b/docs/manuals/characterization/stark_experiment.rst
@@ -112,9 +112,13 @@ drive port of the qubit.
 In a typical IBM device using the cross-resonance drive architecture,
 such channel can be identified with your backend as follows:
 
+.. note::
+    This tutorial requires the :mod:`qiskit_ibm_runtime` package to model a
+    backend.  You can install it with ``python -m pip install qiskit-ibm-runtime``.
+
 .. jupyter-execute::
 
-    from qiskit.providers.fake_provider import FakeHanoiV2
+    from qiskit_ibm_runtime.fake_provider import FakeHanoiV2
 
     backend = FakeHanoiV2()
     qubit = 0
@@ -143,7 +147,7 @@ by a variant of the Hahn-echo pulse sequence [5]_.
 
     from qiskit_experiments.library import StarkRamseyXY
     from qiskit import schedule, pulse
-    from qiskit.providers.fake_provider import FakeHanoi
+    from qiskit_ibm_runtime.fake_provider import FakeHanoi
     from qiskit.visualization.pulse_v2 import IQXSimple
 
     backend = FakeHanoi()
diff --git a/docs/manuals/characterization/t1.rst b/docs/manuals/characterization/t1.rst
index 8a926e9254..82a5801ebe 100644
--- a/docs/manuals/characterization/t1.rst
+++ b/docs/manuals/characterization/t1.rst
@@ -30,8 +30,9 @@ The following code demonstrates a basic run of a :math:`T_1` experiment
 for qubit 0.
 
 .. note::
-    This manual requires the :mod:`qiskit_aer` package to run simulations.
-    You can install it with ``python -m pip install qiskit-aer``.
+    This tutorial requires the :mod:`qiskit_aer` and :mod:`qiskit_ibm_runtime`
+    packages to run simulations.  You can install them with ``python -m pip
+    install qiskit-aer qiskit-ibm-runtime``.
 
 .. jupyter-execute::
 
@@ -42,9 +43,9 @@ for qubit 0.
     from qiskit_experiments.library.characterization.analysis.t1_analysis import T1KerneledAnalysis
 
     # A T1 simulator
-    from qiskit.providers.fake_provider import FakePerth
     from qiskit_aer import AerSimulator
     from qiskit_aer.noise import NoiseModel
+    from qiskit_ibm_runtime.fake_provider import FakePerth
 
     # A kerneled data simulator
     from qiskit_experiments.test.mock_iq_backend import MockIQBackend
diff --git a/docs/manuals/characterization/t2ramsey.rst b/docs/manuals/characterization/t2ramsey.rst
index 6e404d9be4..f7991572e4 100644
--- a/docs/manuals/characterization/t2ramsey.rst
+++ b/docs/manuals/characterization/t2ramsey.rst
@@ -58,13 +58,14 @@ We run the experiment on a simulated backend using Qiskit Aer with a
 pure T1/T2 relaxation noise model.
 
 .. note::
-    This manual requires the :mod:`qiskit_aer` package to run simulations.
-    You can install it with ``python -m pip install qiskit-aer``.
+    This tutorial requires the :mod:`qiskit_aer` and :mod:`qiskit_ibm_runtime`
+    packages to run simulations.  You can install them with ``python -m pip
+    install qiskit-aer qiskit-ibm-runtime``.
 
 .. jupyter-execute::
 
     # A T1 simulator
-    from qiskit.providers.fake_provider import FakePerth
+    from qiskit_ibm_runtime.fake_provider import FakePerth
     from qiskit_aer import AerSimulator
     from qiskit_aer.noise import NoiseModel
     
diff --git a/docs/manuals/characterization/tphi.rst b/docs/manuals/characterization/tphi.rst
index 3e0f25503e..245614af31 100644
--- a/docs/manuals/characterization/tphi.rst
+++ b/docs/manuals/characterization/tphi.rst
@@ -21,8 +21,9 @@ From the :math:`T_1` and :math:`T_2` estimates, we compute the results for
 :math:`T_\varphi.`
 
 .. note::
-    This manual requires the :mod:`qiskit_aer` package to run simulations.
-    You can install it with ``python -m pip install qiskit-aer``.
+    This tutorial requires the :mod:`qiskit_aer` and :mod:`qiskit_ibm_runtime`
+    packages to run simulations.  You can install them with ``python -m pip
+    install qiskit-aer qiskit-ibm-runtime``.
 
 .. jupyter-execute::
 
@@ -31,9 +32,9 @@ From the :math:`T_1` and :math:`T_2` estimates, we compute the results for
     from qiskit_experiments.library.characterization import Tphi
 
     # An Aer simulator
-    from qiskit.providers.fake_provider import FakePerth
     from qiskit_aer import AerSimulator
     from qiskit_aer.noise import NoiseModel
+    from qiskit_ibm_runtime.fake_provider import FakePerth
     
     # Create a pure relaxation noise model for AerSimulator
     noise_model = NoiseModel.from_backend(
diff --git a/docs/manuals/measurement/readout_mitigation.rst b/docs/manuals/measurement/readout_mitigation.rst
index 0fed52b920..8fa861a4ff 100644
--- a/docs/manuals/measurement/readout_mitigation.rst
+++ b/docs/manuals/measurement/readout_mitigation.rst
@@ -31,8 +31,9 @@ This notebook demonstrates the usage of both the local and correlated
 experiments to generate the corresponding mitigators.
 
 .. note::
-    This manual requires the :mod:`qiskit_aer` package to run simulations.
-    You can install it with ``python -m pip install qiskit-aer``.
+    This tutorial requires the :mod:`qiskit_aer` and :mod:`qiskit_ibm_runtime`
+    packages to run simulations.  You can install them with ``python -m pip
+    install qiskit-aer qiskit-ibm-runtime``.
 
 .. jupyter-execute::
 
@@ -43,7 +44,7 @@ experiments to generate the corresponding mitigators.
     from qiskit_experiments.library import LocalReadoutError, CorrelatedReadoutError
 
     from qiskit_aer import AerSimulator
-    from qiskit.providers.fake_provider import FakePerth
+    from qiskit_ibm_runtime.fake_provider import FakePerth
 
     from qiskit.result.mitigation.utils import (
         expval_with_stddev,
diff --git a/docs/manuals/measurement/restless_measurements.rst b/docs/manuals/measurement/restless_measurements.rst
index c62fe75d27..fbde9c7a3e 100644
--- a/docs/manuals/measurement/restless_measurements.rst
+++ b/docs/manuals/measurement/restless_measurements.rst
@@ -1,218 +1,223 @@
-Restless Measurements
-=====================
-
-When running circuits, the qubits are typically reset to the ground state after
-each measurement to ensure that the next circuit has a well-defined initial state.
-This can be done passively by waiting several :math:`T_1`-times so that qubits in
-the excited state decay to :math:`\left\vert0\right\rangle`. Since :math:`T_1`-times
-are continuously improving (they already increased beyond :math:`100\,\mu s`), this
-initialization procedure is now inefficient. This makes active reset necessary.
-Active qubit reset, as employed by IBM Quantum systems is more efficient and saves
-time but also lasts a few microseconds (between :math:`3` and :math:`5\,\mu s`).
-Furthermore, a delay, typically lasting :math:`250\,\mu s`, after the reset
-operation is often necessary to ensure a high initialization quality.
-However, for several types of characterization and calibration experiments we can
-avoid qubit reset by post-processing the measurement outcomes and continue directly
-with the next circuit after an optional short delay, even if the qubit was measured
-in the excited state. Foregoing qubit reset is the main idea behind restless measurements.
-
-The IBM Quantum systems have dynamical repetition delays enabled. We can thus choose
-the delay between the execution of two quantum circuits. This delay
-typically ranges from :math:`0` to :math:`500\,\mu s` depending on the system.
-The default value for most devices is :math:`250\,\mu s`. Restless measurements
-set this delay to small values such as :math:`1\,\mu s` or lower. Note that sometimes
-the measurement instructions already contain a delay after the measurement pulse to
-allow the readout resonator to depopulate.
-
-When the qubit is not reset it will either be in the :math:`\left\vert0\right\rangle`
-or in the :math:`\left\vert1\right\rangle` state when the next circuit starts.
-Therefore, the measured outcomes of the restless experiments require post-processing.
-The following example, taken from Ref. [1]_, illustrates what happens to the single
-measurement outcomes represented as complex numbers in the IQ plane in a restless
-setting. Here, we run three circuits with an identity gate and three circuits with
-an :math:`X` gate, each followed by a measurement. The numbers in the IQ shots indicate the
-order in which the shots were acquired. The IQ plane on the left shows the single
-measurement shots gathered when the qubits are reset. Here, the blue and red points,
-corresponding to measurements following the :math:`Id` and :math:`X` gates, are associated with the
-:math:`\left\vert0\right\rangle` and :math:`\left\vert1\right\rangle` states,
-respectively.
-By contrast, with restless measurements the qubit is not reset after a
-measurement. As one can see in the IQ plane on the right the single measurement
-outcomes of the :math:`Id`` and :math:`X` circuits no longer match with the
-:math:`\left\vert0\right\rangle` and :math:`\left\vert1\right\rangle` states,
-respectively. This is why restless measurements need special post-processing.
-
-.. image:: restless_shots.png
-   :width: 600
-
-Enabling restless measurements
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-In Qiskit Experiments, the experiments that support restless measurements
-have a special method :meth:`~.RestlessMixin.enable_restless` to set the restless run options 
-and define the data processor that will process the measured data.
-If you are an experiment developer, you can add the :class:`.RestlessMixin`
-to your experiment class to add support for restless measurements.
-Here, we will show how to activate restless measurements using
-a fake backend and a rough DRAG experiment. Note however, that you will not
-observe any meaningful outcomes with fake backends since the circuit simulator
-they use always starts with the qubits in the ground state.
-
-.. jupyter-execute::
-
-    from qiskit_experiments.library import RoughDragCal
-    from qiskit_experiments.calibration_management import (
-        Calibrations,
-        FixedFrequencyTransmon,
-    )
-    from qiskit_experiments.data_processing.data_processor import DataProcessor
-    from qiskit.providers.fake_provider import FakePerth
-
-    # replace this lines with an IBM Quantum backend to run the experiment.
-    backend = FakePerth()
-    cals = Calibrations.from_backend(backend, libraries=[FixedFrequencyTransmon()])
-
-    # Define the experiment
-    qubit = 2
-    cal_drag = RoughDragCal((qubit,), cals, schedule_name='sx', backend=backend)
-
-    # Enable restless measurements by setting the run options and data processor
-    cal_drag.enable_restless(rep_delay=1e-6)
-    
-    print(cal_drag.analysis.options.data_processor)
-    print(cal_drag.run_options)
-
-As you can see, a restless data processor is automatically chosen for the experiment. This
-data processor post-processes the restless measured shots according to the order in which
-they were acquired. Furthermore, the appropriate run options are also set. Note that
-these run options might be unique to IBM Quantum providers. Therefore, execute may fail
-on non-IBM Quantum providers if the required options are not supported.
-
-After calling :meth:`~.RestlessMixin.enable_restless` the experiment is ready to be run
-in a restless mode. With a hardware backend, this would be done by calling the
-:meth:`~.BaseExperiment.run` method:
-
-.. jupyter-input::
-
-    drag_data_restless = cal_drag.run()
-
-As shown by the example, the code is identical to running a normal experiment aside
-from a call to the method :meth:`~.RestlessMixin.enable_restless`. Note that you can also choose to keep
-the standard data processor by providing it to the analysis options and telling
-:meth:`~.RestlessMixin.enable_restless` not to override the data processor.
-
-.. jupyter-execute::
-
-    from qiskit_experiments.data_processing import (
-        DataProcessor,
-        Probability,
-    )
-
-    # define a standard data processor.
-    standard_processor = DataProcessor("counts", [Probability("1")])
-
-    cal_drag = RoughDragCal((qubit,), cals, schedule_name='sx', backend=backend)
-    cal_drag.analysis.set_options(data_processor=standard_processor)
-
-    # enable restless mode and set override_processor_by_restless to False.
-    cal_drag.enable_restless(rep_delay=1e-6, override_processor_by_restless=False)
-
-If you run the experiment in this setting you will see that the data is often
-unusable which illustrates the importance of the data processing. As detailed
-in Ref. [2]_, restless measurements can be done with a wide variety
-of experiments such as fine amplitude and drag error amplifying gate sequences
-as well as randomized benchmarking.
-
-Calculating restless quantum processor speed-ups
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-Following Ref. [2]_, we can compare the time spent by the quantum processor executing
-restless and standard jobs. This allows us to compute the effective speed-up we gain
-when performing restless experiments. Note that we do not consider any classical
-run-time contributions such as runtime-compilation or data transfer times [3]_.
-The time to run :math:`K` circuits and gather :math:`N` shots for each
-circuit is
-
-.. math::
-
-    \tau^{(x)} = NK\left(\tau^{(x)}_\text{reset}+\tau^{(x)}_\text{delay}+
-    \langle{\tau}_\text{circ}\rangle+\tau_\text{meas}\right),
-
-where :math:`\tau^{(x)}_\text{reset}` and :math:`\tau^{(x)}_\text{delay}`
-are the reset and post measurement delay times, respectively. The superscript
-:math:`(x)` indicates restless :math:`(r)` or standard :math:`(s)` measurements.
-The average duration of all :math:`K` circuits in an experiment is
-:math:`\langle{\tau}_\text{circ}\rangle=K^{-1}\sum_{k=1}^{K} \tau_{\text{circ},k}`
-where :math:`\tau_{\text{circ},k}` is the duration of only the gates in circuit
-:math:`k`. We therefore compute the quantum processor speed-up of restless
-measurements as :math:`\tau^{(\text{s})}/\tau^{(\text{r})}` which is independent
-of the number of circuits and shots.
-
-We approximate the standard reset time in IBM Quantum backends by
-:math:`\tau^{(s)}_\text{reset} = 4\,\mu s` whereas :math:`\tau^{(r)}_\text{reset} = 0\,\mu s`
-since we do not reset the qubit in a restless experiment. By default, the repetition delay is
-:math:`\tau^{(s)}_\text{delay} = 250\,\mu s`. For our restless experiments we
-set :math:`\tau^{(r)}_\text{delay} = 1\,\mu s`. These speed-ups can be evaluated
-using the code below.
-
-.. jupyter-execute::
-
-    from qiskit import schedule, transpile
-    from qiskit_experiments.framework import BackendData
-
-    dt = BackendData(backend).dt
-    inst_map = backend.instruction_schedule_map
-    meas_length = inst_map.get("measure", (qubit,)).duration * dt
-
-    # Compute the average duration of all circuits
-    # Remove measurement instructions
-    circuits = []
-    for qc in cal_drag.circuits():
-        qc.remove_final_measurements(inplace=True)
-        circuits.append(qc)
-
-    # Schedule the circuits to obtain the duration of all the gates
-    executed_circs = transpile(
-        circuits,
-        backend,
-        initial_layout=[qubit],
-        scheduling_method="alap",
-        **cal_drag.transpile_options.__dict__,
-    )
-    durations = [c.duration for c in executed_circs]
-
-    tau = sum(durations) * dt / (len(durations))
-
-    n_circs = len(cal_drag.circuits())
-    # can be obtained from backend.default_rep_delay on a backend from qiskit-ibm-provider
-
-    delay_s = 0.0025
-    delay_r = 1e-6  # restless delay
-    reset = 4e-6  # Estimated reset duration
-    speed_up = (meas_length + reset + delay_s + tau) / (meas_length + delay_r + tau)
-    print(f"The QPU will spend {speed_up:.1f}x less time running restless Drag.")
-
-The example above is applicable to other experiments and shows that restless
-measurements can greatly speed-up characterization and calibration tasks.
-
-References
-----------
-
-.. [1] Max Werninghaus, Daniel J. Egger, Stefan Filipp, High-speed calibration and
-    characterization of superconducting quantum processors without qubit reset,
-    PRX Quantum **2**, 020324 (2021). https://arxiv.org/abs/2010.06576.
-
-.. [2] Caroline Tornow, Naoki Kanazawa, William E. Shanks, Daniel J. Egger,
-    Minimum quantum run-time characterization and calibration via restless
-    measurements with dynamic repetition rates, Physics Review Applied **17**,
-    064061 (2022). https://arxiv.org/abs/2202.06981.
-
-.. [3] Andrew Wack, Hanhee Paik, Ali Javadi-Abhari, Petar Jurcevic, Ismael Faro,
-    Jay M. Gambetta, Blake R. Johnson, Quality, Speed, and Scale: three key
-    attributes to measure the performance of near-term quantum computers,
-    https://arxiv.org/abs/2110.14108.
-
-See also
---------
-
-* API documentation: :mod:`~qiskit_experiments.framework.RestlessMixin`
+Restless Measurements
+=====================
+
+When running circuits, the qubits are typically reset to the ground state after
+each measurement to ensure that the next circuit has a well-defined initial state.
+This can be done passively by waiting several :math:`T_1`-times so that qubits in
+the excited state decay to :math:`\left\vert0\right\rangle`. Since :math:`T_1`-times
+are continuously improving (they already increased beyond :math:`100\,\mu s`), this
+initialization procedure is now inefficient. This makes active reset necessary.
+Active qubit reset, as employed by IBM Quantum systems is more efficient and saves
+time but also lasts a few microseconds (between :math:`3` and :math:`5\,\mu s`).
+Furthermore, a delay, typically lasting :math:`250\,\mu s`, after the reset
+operation is often necessary to ensure a high initialization quality.
+However, for several types of characterization and calibration experiments we can
+avoid qubit reset by post-processing the measurement outcomes and continue directly
+with the next circuit after an optional short delay, even if the qubit was measured
+in the excited state. Foregoing qubit reset is the main idea behind restless measurements.
+
+The IBM Quantum systems have dynamical repetition delays enabled. We can thus choose
+the delay between the execution of two quantum circuits. This delay
+typically ranges from :math:`0` to :math:`500\,\mu s` depending on the system.
+The default value for most devices is :math:`250\,\mu s`. Restless measurements
+set this delay to small values such as :math:`1\,\mu s` or lower. Note that sometimes
+the measurement instructions already contain a delay after the measurement pulse to
+allow the readout resonator to depopulate.
+
+When the qubit is not reset it will either be in the :math:`\left\vert0\right\rangle`
+or in the :math:`\left\vert1\right\rangle` state when the next circuit starts.
+Therefore, the measured outcomes of the restless experiments require post-processing.
+The following example, taken from Ref. [1]_, illustrates what happens to the single
+measurement outcomes represented as complex numbers in the IQ plane in a restless
+setting. Here, we run three circuits with an identity gate and three circuits with
+an :math:`X` gate, each followed by a measurement. The numbers in the IQ shots indicate the
+order in which the shots were acquired. The IQ plane on the left shows the single
+measurement shots gathered when the qubits are reset. Here, the blue and red points,
+corresponding to measurements following the :math:`Id` and :math:`X` gates, are associated with the
+:math:`\left\vert0\right\rangle` and :math:`\left\vert1\right\rangle` states,
+respectively.
+By contrast, with restless measurements the qubit is not reset after a
+measurement. As one can see in the IQ plane on the right the single measurement
+outcomes of the :math:`Id`` and :math:`X` circuits no longer match with the
+:math:`\left\vert0\right\rangle` and :math:`\left\vert1\right\rangle` states,
+respectively. This is why restless measurements need special post-processing.
+
+.. image:: restless_shots.png
+   :width: 600
+
+Enabling restless measurements
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+In Qiskit Experiments, the experiments that support restless measurements
+have a special method :meth:`~.RestlessMixin.enable_restless` to set the restless run options
+and define the data processor that will process the measured data.
+If you are an experiment developer, you can add the :class:`.RestlessMixin`
+to your experiment class to add support for restless measurements.
+Here, we will show how to activate restless measurements using
+a fake backend and a rough DRAG experiment. Note however, that you will not
+observe any meaningful outcomes with fake backends since the circuit simulator
+they use always starts with the qubits in the ground state.
+
+.. note::
+    This tutorial requires the :mod:`qiskit_ibm_runtime` package to model a
+    backend.  You can install it with ``python -m pip install qiskit-ibm-runtime``.
+
+.. jupyter-execute::
+
+    from qiskit_ibm_runtime.fake_provider import FakePerth
+
+    from qiskit_experiments.library import RoughDragCal
+    from qiskit_experiments.calibration_management import (
+        Calibrations,
+        FixedFrequencyTransmon,
+    )
+    from qiskit_experiments.data_processing.data_processor import DataProcessor
+
+    # replace this lines with an IBM Quantum backend to run the experiment.
+    backend = FakePerth()
+    cals = Calibrations.from_backend(backend, libraries=[FixedFrequencyTransmon()])
+
+    # Define the experiment
+    qubit = 2
+    cal_drag = RoughDragCal((qubit,), cals, schedule_name='sx', backend=backend)
+
+    # Enable restless measurements by setting the run options and data processor
+    cal_drag.enable_restless(rep_delay=1e-6)
+
+    print(cal_drag.analysis.options.data_processor)
+    print(cal_drag.run_options)
+
+As you can see, a restless data processor is automatically chosen for the experiment. This
+data processor post-processes the restless measured shots according to the order in which
+they were acquired. Furthermore, the appropriate run options are also set. Note that
+these run options might be unique to IBM Quantum providers. Therefore, execute may fail
+on non-IBM Quantum providers if the required options are not supported.
+
+After calling :meth:`~.RestlessMixin.enable_restless` the experiment is ready to be run
+in a restless mode. With a hardware backend, this would be done by calling the
+:meth:`~.BaseExperiment.run` method:
+
+.. jupyter-input::
+
+    drag_data_restless = cal_drag.run()
+
+As shown by the example, the code is identical to running a normal experiment aside
+from a call to the method :meth:`~.RestlessMixin.enable_restless`. Note that you can also choose to keep
+the standard data processor by providing it to the analysis options and telling
+:meth:`~.RestlessMixin.enable_restless` not to override the data processor.
+
+.. jupyter-execute::
+
+    from qiskit_experiments.data_processing import (
+        DataProcessor,
+        Probability,
+    )
+
+    # define a standard data processor.
+    standard_processor = DataProcessor("counts", [Probability("1")])
+
+    cal_drag = RoughDragCal((qubit,), cals, schedule_name='sx', backend=backend)
+    cal_drag.analysis.set_options(data_processor=standard_processor)
+
+    # enable restless mode and set override_processor_by_restless to False.
+    cal_drag.enable_restless(rep_delay=1e-6, override_processor_by_restless=False)
+
+If you run the experiment in this setting you will see that the data is often
+unusable which illustrates the importance of the data processing. As detailed
+in Ref. [2]_, restless measurements can be done with a wide variety
+of experiments such as fine amplitude and drag error amplifying gate sequences
+as well as randomized benchmarking.
+
+Calculating restless quantum processor speed-ups
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Following Ref. [2]_, we can compare the time spent by the quantum processor executing
+restless and standard jobs. This allows us to compute the effective speed-up we gain
+when performing restless experiments. Note that we do not consider any classical
+run-time contributions such as runtime-compilation or data transfer times [3]_.
+The time to run :math:`K` circuits and gather :math:`N` shots for each
+circuit is
+
+.. math::
+
+    \tau^{(x)} = NK\left(\tau^{(x)}_\text{reset}+\tau^{(x)}_\text{delay}+
+    \langle{\tau}_\text{circ}\rangle+\tau_\text{meas}\right),
+
+where :math:`\tau^{(x)}_\text{reset}` and :math:`\tau^{(x)}_\text{delay}`
+are the reset and post measurement delay times, respectively. The superscript
+:math:`(x)` indicates restless :math:`(r)` or standard :math:`(s)` measurements.
+The average duration of all :math:`K` circuits in an experiment is
+:math:`\langle{\tau}_\text{circ}\rangle=K^{-1}\sum_{k=1}^{K} \tau_{\text{circ},k}`
+where :math:`\tau_{\text{circ},k}` is the duration of only the gates in circuit
+:math:`k`. We therefore compute the quantum processor speed-up of restless
+measurements as :math:`\tau^{(\text{s})}/\tau^{(\text{r})}` which is independent
+of the number of circuits and shots.
+
+We approximate the standard reset time in IBM Quantum backends by
+:math:`\tau^{(s)}_\text{reset} = 4\,\mu s` whereas :math:`\tau^{(r)}_\text{reset} = 0\,\mu s`
+since we do not reset the qubit in a restless experiment. By default, the repetition delay is
+:math:`\tau^{(s)}_\text{delay} = 250\,\mu s`. For our restless experiments we
+set :math:`\tau^{(r)}_\text{delay} = 1\,\mu s`. These speed-ups can be evaluated
+using the code below.
+
+.. jupyter-execute::
+
+    from qiskit import schedule, transpile
+    from qiskit_experiments.framework import BackendData
+
+    dt = BackendData(backend).dt
+    inst_map = backend.instruction_schedule_map
+    meas_length = inst_map.get("measure", (qubit,)).duration * dt
+
+    # Compute the average duration of all circuits
+    # Remove measurement instructions
+    circuits = []
+    for qc in cal_drag.circuits():
+        qc.remove_final_measurements(inplace=True)
+        circuits.append(qc)
+
+    # Schedule the circuits to obtain the duration of all the gates
+    executed_circs = transpile(
+        circuits,
+        backend,
+        initial_layout=[qubit],
+        scheduling_method="alap",
+        **cal_drag.transpile_options.__dict__,
+    )
+    durations = [c.duration for c in executed_circs]
+
+    tau = sum(durations) * dt / (len(durations))
+
+    n_circs = len(cal_drag.circuits())
+    # can be obtained from backend.default_rep_delay on a backend from qiskit-ibm-provider
+
+    delay_s = 0.0025
+    delay_r = 1e-6  # restless delay
+    reset = 4e-6  # Estimated reset duration
+    speed_up = (meas_length + reset + delay_s + tau) / (meas_length + delay_r + tau)
+    print(f"The QPU will spend {speed_up:.1f}x less time running restless Drag.")
+
+The example above is applicable to other experiments and shows that restless
+measurements can greatly speed-up characterization and calibration tasks.
+
+References
+----------
+
+.. [1] Max Werninghaus, Daniel J. Egger, Stefan Filipp, High-speed calibration and
+    characterization of superconducting quantum processors without qubit reset,
+    PRX Quantum **2**, 020324 (2021). https://arxiv.org/abs/2010.06576.
+
+.. [2] Caroline Tornow, Naoki Kanazawa, William E. Shanks, Daniel J. Egger,
+    Minimum quantum run-time characterization and calibration via restless
+    measurements with dynamic repetition rates, Physics Review Applied **17**,
+    064061 (2022). https://arxiv.org/abs/2202.06981.
+
+.. [3] Andrew Wack, Hanhee Paik, Ali Javadi-Abhari, Petar Jurcevic, Ismael Faro,
+    Jay M. Gambetta, Blake R. Johnson, Quality, Speed, and Scale: three key
+    attributes to measure the performance of near-term quantum computers,
+    https://arxiv.org/abs/2110.14108.
+
+See also
+--------
+
+* API documentation: :mod:`~qiskit_experiments.framework.RestlessMixin`
diff --git a/docs/manuals/verification/quantum_volume.rst b/docs/manuals/verification/quantum_volume.rst
index 5c06e5f1e5..0497800881 100644
--- a/docs/manuals/verification/quantum_volume.rst
+++ b/docs/manuals/verification/quantum_volume.rst
@@ -25,17 +25,19 @@ probability` > 2/3 with confidence level > 0.977 (corresponding to
 z_value = 2), and at least 100 trials have been ran.
 
 .. note::
-    This manual requires the :mod:`qiskit_aer` package to run simulations.
-    You can install it with ``python -m pip install qiskit-aer``.
+    This tutorial requires the :mod:`qiskit_aer` and :mod:`qiskit_ibm_runtime`
+    packages to run simulations.  You can install them with ``python -m pip
+    install qiskit-aer qiskit-ibm-runtime``.
 
 .. jupyter-execute::
 
     from qiskit_experiments.framework import BatchExperiment
     from qiskit_experiments.library import QuantumVolume
-    from qiskit_aer import AerSimulator
     
     # For simulation
-    from qiskit.providers.fake_provider import FakeSydneyV2
+    from qiskit import Aer
+    from qiskit_aer import AerSimulator
+    from qiskit_ibm_runtime.fake_provider import FakeSydneyV2
     
     backend = AerSimulator.from_backend(FakeSydneyV2())
 
diff --git a/docs/manuals/verification/randomized_benchmarking.rst b/docs/manuals/verification/randomized_benchmarking.rst
index 111e3f5ddc..50e7620356 100644
--- a/docs/manuals/verification/randomized_benchmarking.rst
+++ b/docs/manuals/verification/randomized_benchmarking.rst
@@ -11,6 +11,11 @@ error estimates for the quantum device, by calculating the Error Per Clifford. S
 <https://github.com/Qiskit/textbook/blob/main/notebooks/quantum-hardware/randomized-benchmarking.ipynb>`__ for an
 explanation on the RB method, which is based on Refs. [1]_ [2]_.
 
+.. note::
+    This tutorial requires the :mod:`qiskit_aer` and :mod:`qiskit_ibm_runtime`
+    packages to run simulations.  You can install them with ``python -m pip
+    install qiskit-aer qiskit-ibm-runtime``.
+
 .. jupyter-execute::
 
     import numpy as np
@@ -20,7 +25,7 @@ explanation on the RB method, which is based on Refs. [1]_ [2]_.
     
     # For simulation
     from qiskit_aer import AerSimulator
-    from qiskit.providers.fake_provider import FakePerth
+    from qiskit_ibm_runtime.fake_provider import FakePerth
     
     backend = AerSimulator.from_backend(FakePerth())
 
diff --git a/docs/manuals/verification/state_tomography.rst b/docs/manuals/verification/state_tomography.rst
index f207365756..dd972041dc 100644
--- a/docs/manuals/verification/state_tomography.rst
+++ b/docs/manuals/verification/state_tomography.rst
@@ -8,15 +8,16 @@ of a quantum state by preparing the state many times and measuring them in a tom
 complete basis of measurement operators.
 
 .. note::
-    This tutorial requires the :mod:`qiskit_aer` package for simulations.
-    You can install it with ``python -m pip install qiskit-aer``.
+    This tutorial requires the :mod:`qiskit_aer` and :mod:`qiskit_ibm_runtime`
+    packages to run simulations.  You can install them with ``python -m pip
+    install qiskit-aer qiskit-ibm-runtime``.
 
 We first initialize a simulator to run the experiments on.
 
 .. jupyter-execute::
 
     from qiskit_aer import AerSimulator
-    from qiskit.providers.fake_provider import FakePerth
+    from qiskit_ibm_runtime.fake_provider import FakePerth
     
     backend = AerSimulator.from_backend(FakePerth())
 
diff --git a/docs/tutorials/getting_started.rst b/docs/tutorials/getting_started.rst
index d702f0d72b..7ca6e32567 100644
--- a/docs/tutorials/getting_started.rst
+++ b/docs/tutorials/getting_started.rst
@@ -54,16 +54,17 @@ the experiment from the Qiskit Experiments library:
     from qiskit_experiments.library import T1
 
 Experiments must be run on a backend. We're going to use a simulator,
-:class:`~qiskit.providers.fake_provider.FakePerth`, for this example, but you can use any
+:class:`~qiskit_ibm_runtime.fake_provider.FakePerth`, for this example, but you can use any
 backend, real or simulated, that you can access through Qiskit.
 
 .. note::
-    This tutorial requires the :mod:`qiskit_aer` package to run simulations.
-    You can install it with ``python -m pip install qiskit-aer``.
+    This tutorial requires the :mod:`qiskit_aer` and :mod:`qiskit_ibm_runtime`
+    packages to run simulations.  You can install them with ``python -m pip
+    install qiskit-aer qiskit-ibm-runtime``.
 
 .. jupyter-execute::
 
-    from qiskit.providers.fake_provider import FakePerth
+    from qiskit_ibm_runtime.fake_provider import FakePerth
     from qiskit_aer import AerSimulator
 
     backend = AerSimulator.from_backend(FakePerth())
@@ -419,4 +420,4 @@ Use the `broadcast` parameter to set analysis options to each of the child exper
 
 If the child experiment inherits from :class:`.CompositeExperiment` (such as :class:`.ParallelExperiment`
 and :class:`.BatchExperiment` classes), this process will continue to work recursively.
-In this instance, the analysis will not generate a figure for the child experiment after the analysis.
\ No newline at end of file
+In this instance, the analysis will not generate a figure for the child experiment after the analysis.
diff --git a/docs/tutorials/visualization.rst b/docs/tutorials/visualization.rst
index 1c8b5ef05d..5dc5e39483 100644
--- a/docs/tutorials/visualization.rst
+++ b/docs/tutorials/visualization.rst
@@ -34,8 +34,9 @@ Generating and customizing a figure using a plotter
 First, we display the default figure from a :class:`.Rabi` experiment as a starting point:
 
 .. note::
-    This tutorial requires the :mod:`qiskit_dynamics` package to run simulations.
-    You can install it with ``python -m pip install qiskit-dynamics``.
+    This tutorial requires the :mod:`qiskit_dynamics`, :mod:`qiskit_aer`, and
+    :mod:`qiskit_ibm_runtime` packages to run simulations.  You can install them
+    with ``python -m pip install qiskit-dynamics qiskit-aer qiskit-ibm-runtime``.
 
 .. jupyter-execute::
 
@@ -191,8 +192,8 @@ until now in this tutorial:
 .. jupyter-execute::
 
    from qiskit_experiments.library import T1
-   from qiskit.providers.fake_provider import FakePerth
    from qiskit_aer import AerSimulator
+   from qiskit_ibm_runtime.fake_provider import FakePerth
 
    backend = AerSimulator.from_backend(FakePerth())
 
diff --git a/qiskit_experiments/test/mock_iq_backend.py b/qiskit_experiments/test/mock_iq_backend.py
index 09094db37a..8626af8aac 100644
--- a/qiskit_experiments/test/mock_iq_backend.py
+++ b/qiskit_experiments/test/mock_iq_backend.py
@@ -19,9 +19,9 @@
 from qiskit.circuit.library import XGate, SXGate
 from qiskit.result import Result
 from qiskit.providers.fake_provider import FakeOpenPulse2Q
-from qiskit.providers.fake_provider.fake_backend import FakeBackendV2
-
 from qiskit.qobj.utils import MeasLevel
+from qiskit_ibm_runtime.fake_provider.fake_backend import FakeBackendV2
+
 from qiskit_experiments.exceptions import QiskitError
 from qiskit_experiments.framework import Options
 from qiskit_experiments.test.utils import FakeJob
diff --git a/requirements-dev.txt b/requirements-dev.txt
index 934285c6eb..3594ba0434 100644
--- a/requirements-dev.txt
+++ b/requirements-dev.txt
@@ -1,3 +1,4 @@
+qiskit==0.45.3
 # Linters
 black~=22.0
 pylint~=3.0.2
@@ -12,6 +13,7 @@ testtools
 
 # Extra dependencies for tests/documentation code
 multimethod
+qiskit-ibm-runtime>=0.18
 
 # Documentation tools
 arxiv
diff --git a/requirements.txt b/requirements.txt
index 2bda6ad888..23690b345a 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -1,6 +1,6 @@
 numpy>=1.17
 scipy>=1.4
-qiskit>=0.45
+qiskit==0.45.3  # Temporary
 qiskit-ibm-experiment>=0.3.4
 matplotlib>=3.4
 uncertainties
diff --git a/test/calibration/test_calibrations.py b/test/calibration/test_calibrations.py
index 3b99cd7668..68c6a5de71 100644
--- a/test/calibration/test_calibrations.py
+++ b/test/calibration/test_calibrations.py
@@ -36,8 +36,8 @@
 from qiskit.circuit.library import CXGate, XGate
 from qiskit.pulse.transforms import inline_subroutines, block_to_schedule
 from qiskit.providers import BackendV2, Options
-from qiskit.providers.fake_provider import FakeArmonkV2, FakeBelemV2
 from qiskit.transpiler import Target
+from qiskit_ibm_runtime.fake_provider import FakeArmonkV2, FakeBelemV2
 
 from qiskit_experiments.framework import BackendData
 from qiskit_experiments.calibration_management.calibrations import Calibrations, ParameterKey
diff --git a/test/calibration/test_update_library.py b/test/calibration/test_update_library.py
index 0cc8186dfc..027b21b511 100644
--- a/test/calibration/test_update_library.py
+++ b/test/calibration/test_update_library.py
@@ -14,8 +14,8 @@
 from test.base import QiskitExperimentsTestCase
 import numpy as np
 
-from qiskit.providers.fake_provider import FakeAthensV2
 from qiskit.qobj.utils import MeasLevel
+from qiskit_ibm_runtime.fake_provider import FakeAthensV2
 
 from qiskit_experiments.framework import BackendData
 from qiskit_experiments.library import QubitSpectroscopy
diff --git a/test/database_service/test_db_experiment_data.py b/test/database_service/test_db_experiment_data.py
index 164e5160f7..57958ae34b 100644
--- a/test/database_service/test_db_experiment_data.py
+++ b/test/database_service/test_db_experiment_data.py
@@ -29,7 +29,7 @@
 import matplotlib.pyplot as plt
 import numpy as np
 
-from qiskit.providers.fake_provider import FakeMelbourneV2
+from qiskit_ibm_runtime.fake_provider import FakeMelbourneV2
 from qiskit.result import Result
 from qiskit.providers import JobV1 as Job
 from qiskit.providers import JobStatus
diff --git a/test/framework/test_backend_timing.py b/test/framework/test_backend_timing.py
index c270795069..a7951b2c21 100644
--- a/test/framework/test_backend_timing.py
+++ b/test/framework/test_backend_timing.py
@@ -16,7 +16,7 @@
 
 from ddt import data, ddt, unpack
 from qiskit import QiskitError
-from qiskit.providers.fake_provider import FakeNairobiV2
+from qiskit_ibm_runtime.fake_provider import FakeNairobiV2
 
 from qiskit_experiments.framework import BackendTiming
 
diff --git a/test/framework/test_framework.py b/test/framework/test_framework.py
index 105a5ffdf5..0d52da1d84 100644
--- a/test/framework/test_framework.py
+++ b/test/framework/test_framework.py
@@ -19,9 +19,10 @@
 import ddt
 
 from qiskit import QuantumCircuit
-from qiskit.providers.fake_provider import FakeVigoV2, FakeJob
+from qiskit.providers.fake_provider import FakeJob
 from qiskit.providers.jobstatus import JobStatus
 from qiskit.exceptions import QiskitError
+from qiskit_ibm_runtime.fake_provider import FakeVigoV2
 
 from qiskit_experiments.exceptions import AnalysisError
 from qiskit_experiments.framework import (
diff --git a/test/library/calibration/test_drag.py b/test/library/calibration/test_drag.py
index b66e735f3b..d870b83d3a 100644
--- a/test/library/calibration/test_drag.py
+++ b/test/library/calibration/test_drag.py
@@ -20,9 +20,9 @@
 from qiskit import pulse
 from qiskit.circuit import Parameter
 from qiskit.exceptions import QiskitError
-from qiskit.providers.fake_provider import FakeWashingtonV2
 from qiskit.pulse import DriveChannel, Drag
 from qiskit.qobj.utils import MeasLevel
+from qiskit_ibm_runtime.fake_provider import FakeWashingtonV2
 
 from qiskit_experiments.library import RoughDrag, RoughDragCal
 from qiskit_experiments.library.characterization.analysis import DragCalAnalysis
diff --git a/test/library/calibration/test_fine_amplitude.py b/test/library/calibration/test_fine_amplitude.py
index 659e0f758a..7608f434b5 100644
--- a/test/library/calibration/test_fine_amplitude.py
+++ b/test/library/calibration/test_fine_amplitude.py
@@ -18,8 +18,8 @@
 from qiskit import pulse
 from qiskit.circuit import Gate
 from qiskit.circuit.library import XGate, SXGate
-from qiskit.providers.fake_provider import FakeArmonkV2
 from qiskit.pulse import DriveChannel, Drag
+from qiskit_ibm_runtime.fake_provider import FakeArmonkV2
 
 from qiskit_experiments.library import (
     FineXAmplitude,
diff --git a/test/library/calibration/test_fine_drag.py b/test/library/calibration/test_fine_drag.py
index 818152c07d..f592008c17 100644
--- a/test/library/calibration/test_fine_drag.py
+++ b/test/library/calibration/test_fine_drag.py
@@ -17,7 +17,7 @@
 
 from qiskit import pulse
 from qiskit.circuit import Gate
-from qiskit.providers.fake_provider import FakeArmonkV2
+from qiskit_ibm_runtime.fake_provider import FakeArmonkV2
 
 from qiskit_experiments.library import FineDrag, FineXDrag, FineDragCal
 from qiskit_experiments.test.mock_iq_backend import MockIQBackend
diff --git a/test/library/calibration/test_fine_frequency.py b/test/library/calibration/test_fine_frequency.py
index 87aa30a175..e6d6f78d1d 100644
--- a/test/library/calibration/test_fine_frequency.py
+++ b/test/library/calibration/test_fine_frequency.py
@@ -17,7 +17,7 @@
 from ddt import ddt, data
 
 from qiskit import pulse
-from qiskit.providers.fake_provider import FakeArmonkV2
+from qiskit_ibm_runtime.fake_provider import FakeArmonkV2
 
 from qiskit_experiments.library import (
     FineFrequency,
diff --git a/test/library/calibration/test_ramsey_xy.py b/test/library/calibration/test_ramsey_xy.py
index 76bd953211..43ba1dbf67 100644
--- a/test/library/calibration/test_ramsey_xy.py
+++ b/test/library/calibration/test_ramsey_xy.py
@@ -16,8 +16,8 @@
 from test.base import QiskitExperimentsTestCase
 
 from ddt import ddt, data, named_data
-from qiskit.providers.fake_provider import FakeArmonkV2
 from qiskit_aer import AerSimulator
+from qiskit_ibm_runtime.fake_provider import FakeArmonkV2
 
 from qiskit_experiments.calibration_management.calibrations import Calibrations
 from qiskit_experiments.calibration_management.basis_gate_library import FixedFrequencyTransmon
diff --git a/test/library/characterization/test_cross_resonance_hamiltonian.py b/test/library/characterization/test_cross_resonance_hamiltonian.py
index 540c0e0dc4..61176cca24 100644
--- a/test/library/characterization/test_cross_resonance_hamiltonian.py
+++ b/test/library/characterization/test_cross_resonance_hamiltonian.py
@@ -19,8 +19,8 @@
 import io
 import numpy as np
 from ddt import ddt, data, unpack
+
 from qiskit import QuantumCircuit, pulse, qpy, quantum_info as qi
-from qiskit.providers.fake_provider import FakeBogotaV2
 
 # TODO: remove old path after we stop supporting the relevant version of Qiskit
 try:
@@ -29,6 +29,8 @@
     from qiskit.extensions.hamiltonian_gate import HamiltonianGate
 
 from qiskit_aer import AerSimulator
+from qiskit_ibm_runtime.fake_provider import FakeBogotaV2
+
 from qiskit_experiments.library.characterization import cr_hamiltonian
 
 
diff --git a/test/library/characterization/test_half_angle.py b/test/library/characterization/test_half_angle.py
index e90c6b6d96..d131fcae99 100644
--- a/test/library/characterization/test_half_angle.py
+++ b/test/library/characterization/test_half_angle.py
@@ -16,8 +16,8 @@
 import copy
 
 from qiskit import pulse, transpile
-from qiskit.providers.fake_provider import FakeAthens
 from qiskit.pulse import InstructionScheduleMap
+from qiskit_ibm_runtime.fake_provider import FakeAthens
 
 from qiskit_experiments.test.mock_iq_backend import MockIQBackend
 from qiskit_experiments.test.mock_iq_helpers import MockIQHalfAngleHelper as HalfAngleHelper
diff --git a/test/library/characterization/test_qubit_spectroscopy.py b/test/library/characterization/test_qubit_spectroscopy.py
index 12ae08a4a5..8dc52bfeb9 100644
--- a/test/library/characterization/test_qubit_spectroscopy.py
+++ b/test/library/characterization/test_qubit_spectroscopy.py
@@ -16,7 +16,7 @@
 
 from qiskit.qobj.utils import MeasLevel
 from qiskit.circuit.library import XGate
-from qiskit.providers.fake_provider import FakeWashingtonV2
+from qiskit_ibm_runtime.fake_provider import FakeWashingtonV2
 from qiskit_experiments.framework import ParallelExperiment
 
 from qiskit_experiments.framework import BackendData
diff --git a/test/library/characterization/test_readout_error.py b/test/library/characterization/test_readout_error.py
index ef4107d413..0e32e75d0e 100644
--- a/test/library/characterization/test_readout_error.py
+++ b/test/library/characterization/test_readout_error.py
@@ -19,9 +19,9 @@
 from test.base import QiskitExperimentsTestCase
 import numpy as np
 from qiskit.quantum_info.operators.predicates import matrix_equal
-from qiskit.providers.fake_provider import FakeParisV2
-from qiskit_ibm_experiment import IBMExperimentService
 from qiskit_aer import AerSimulator
+from qiskit_ibm_experiment import IBMExperimentService
+from qiskit_ibm_runtime.fake_provider import FakeParisV2
 from qiskit_experiments.library.characterization import LocalReadoutError, CorrelatedReadoutError
 from qiskit_experiments.framework import ExperimentData
 from qiskit_experiments.framework import ParallelExperiment
diff --git a/test/library/characterization/test_t1.py b/test/library/characterization/test_t1.py
index 6c49a9130d..475e1d82e9 100644
--- a/test/library/characterization/test_t1.py
+++ b/test/library/characterization/test_t1.py
@@ -16,7 +16,7 @@
 from test.base import QiskitExperimentsTestCase
 import numpy as np
 from qiskit.qobj.utils import MeasLevel
-from qiskit.providers.fake_provider import FakeAthensV2
+from qiskit_ibm_runtime.fake_provider import FakeAthensV2
 from qiskit_experiments.test.noisy_delay_aer_simulator import NoisyDelayAerBackend
 from qiskit_experiments.framework import ExperimentData, ParallelExperiment
 from qiskit_experiments.library import T1
diff --git a/test/library/characterization/test_t2hahn.py b/test/library/characterization/test_t2hahn.py
index 8e3d124e48..6c068a0e75 100644
--- a/test/library/characterization/test_t2hahn.py
+++ b/test/library/characterization/test_t2hahn.py
@@ -18,8 +18,8 @@
 import numpy as np
 from ddt import ddt, data, named_data, unpack
 
-from qiskit.providers.fake_provider import FakeVigoV2
 from qiskit_aer import AerSimulator
+from qiskit_ibm_runtime.fake_provider import FakeVigoV2
 
 from qiskit_experiments.framework import ParallelExperiment
 from qiskit_experiments.library.characterization.t2hahn import T2Hahn
diff --git a/test/library/characterization/test_t2ramsey.py b/test/library/characterization/test_t2ramsey.py
index 599c9a547a..ca6a961981 100644
--- a/test/library/characterization/test_t2ramsey.py
+++ b/test/library/characterization/test_t2ramsey.py
@@ -18,8 +18,8 @@
 import numpy as np
 from ddt import ddt, named_data
 
-from qiskit.providers.fake_provider import FakeVigoV2
 from qiskit_aer import AerSimulator
+from qiskit_ibm_runtime.fake_provider import FakeVigoV2
 
 from qiskit_experiments.framework import ParallelExperiment
 from qiskit_experiments.library import T2Ramsey
diff --git a/test/library/characterization/test_zz_ramsey.py b/test/library/characterization/test_zz_ramsey.py
index b2d45675c8..10b12950a0 100644
--- a/test/library/characterization/test_zz_ramsey.py
+++ b/test/library/characterization/test_zz_ramsey.py
@@ -21,8 +21,8 @@
 from ddt import ddt, idata, named_data, unpack
 
 from qiskit import QuantumCircuit
-from qiskit.providers.fake_provider import FakeVigoV2
 from qiskit_aer import AerSimulator
+from qiskit_ibm_runtime.fake_provider import FakeVigoV2
 
 from qiskit_experiments.library import ZZRamsey
 from qiskit_experiments.test.mock_iq_backend import MockIQBackend
diff --git a/test/library/driven_freq_tuning/test_stark_p1_spect.py b/test/library/driven_freq_tuning/test_stark_p1_spect.py
index 00ddbd5c6f..ed3be9c25b 100644
--- a/test/library/driven_freq_tuning/test_stark_p1_spect.py
+++ b/test/library/driven_freq_tuning/test_stark_p1_spect.py
@@ -19,7 +19,7 @@
 from qiskit import pulse
 from qiskit.circuit import QuantumCircuit, Gate
 from qiskit.providers import QubitProperties
-from qiskit.providers.fake_provider import FakeHanoiV2
+from qiskit_ibm_runtime.fake_provider import FakeHanoiV2
 
 from qiskit_experiments.framework import ExperimentData, AnalysisResultData
 from qiskit_experiments.library import StarkP1Spectroscopy
diff --git a/test/library/driven_freq_tuning/test_stark_ramsey_xy.py b/test/library/driven_freq_tuning/test_stark_ramsey_xy.py
index dd0fd59e4a..3791ce9d7a 100644
--- a/test/library/driven_freq_tuning/test_stark_ramsey_xy.py
+++ b/test/library/driven_freq_tuning/test_stark_ramsey_xy.py
@@ -18,7 +18,7 @@
 from ddt import ddt, unpack, named_data
 
 from qiskit import pulse
-from qiskit.providers.fake_provider import FakeHanoiV2
+from qiskit_ibm_runtime.fake_provider import FakeHanoiV2
 
 from qiskit_experiments.library import StarkRamseyXY, StarkRamseyXYAmpScan
 from qiskit_experiments.library.driven_freq_tuning.ramsey_amp_scan_analysis import (
diff --git a/test/library/randomized_benchmarking/test_interleaved_rb.py b/test/library/randomized_benchmarking/test_interleaved_rb.py
index 1f7b567a23..f01fd6a764 100644
--- a/test/library/randomized_benchmarking/test_interleaved_rb.py
+++ b/test/library/randomized_benchmarking/test_interleaved_rb.py
@@ -20,10 +20,10 @@
 from qiskit.circuit import Delay, QuantumCircuit, Parameter, Gate
 from qiskit.circuit.library import SXGate, CXGate, TGate, CZGate
 from qiskit.exceptions import QiskitError
-from qiskit.providers.fake_provider import FakeManila, FakeManilaV2, FakeWashington
 from qiskit.transpiler import InstructionProperties
 from qiskit_aer import AerSimulator
 from qiskit_aer.noise import NoiseModel, depolarizing_error
+from qiskit_ibm_runtime.fake_provider import FakeManila, FakeManilaV2, FakeWashington
 from qiskit_experiments.library import randomized_benchmarking as rb
 
 
diff --git a/test/library/randomized_benchmarking/test_standard_rb.py b/test/library/randomized_benchmarking/test_standard_rb.py
index fde71cc5e3..2ccd5a47df 100644
--- a/test/library/randomized_benchmarking/test_standard_rb.py
+++ b/test/library/randomized_benchmarking/test_standard_rb.py
@@ -19,10 +19,10 @@
 
 from qiskit.circuit.library import SXGate
 from qiskit.exceptions import QiskitError
-from qiskit.providers.fake_provider import FakeManilaV2
 from qiskit.pulse import Schedule, InstructionScheduleMap
 from qiskit_aer import AerSimulator
 from qiskit_aer.noise import NoiseModel, depolarizing_error
+from qiskit_ibm_runtime.fake_provider import FakeManilaV2
 from qiskit_experiments.framework.composite import ParallelExperiment
 from qiskit_experiments.library import randomized_benchmarking as rb
 
@@ -372,7 +372,7 @@ def test_poor_experiment_result(self):
         This is a special case that fit outcome is very sensitive to initial guess.
         Perhaps generated initial guess is close to a local minima.
         """
-        from qiskit.providers.fake_provider import FakeVigoV2
+        from qiskit_ibm_runtime.fake_provider import FakeVigoV2
 
         backend = FakeVigoV2()
         backend.set_options(seed_simulator=123)