Qiskit Aer Simulator

0.17.2 · active · verified Tue Apr 14

Qiskit Aer is a high-performance simulator backend for Qiskit, designed for fast and accurate simulation of quantum circuits on classical computers. It supports various simulation methods, including statevector, density matrix, extended stabilizer, and matrix product state, along with advanced noise modeling capabilities. The current version is 0.17.2, and it typically releases new versions in sync with major Qiskit releases to maintain compatibility.

Warnings

breaking Qiskit Aer versions prior to 0.17.0 are not fully compatible with Qiskit 2.0 and later. Significant API changes in Qiskit 2.0 require corresponding updates in Qiskit Aer.
Fix: Upgrade `qiskit-aer` to 0.17.0 or newer (e.g., `pip install qiskit-aer>=0.17.0`). Ensure your `qiskit` package is also up-to-date (e.g., `pip install qiskit>=2.0`).
deprecated Directly accessing simulator backends via `Aer.get_backend('aer_simulator')` is less explicit than importing the `AerSimulator` class directly. While still functional, the direct class import is the recommended pattern.
Fix: Prefer `from qiskit_aer import AerSimulator; simulator = AerSimulator()` over `from qiskit_aer import Aer; simulator = Aer.get_backend('aer_simulator')`.
gotcha Qiskit Aer has specific Python version requirements that can change between releases. For example, version 0.16.1 dropped support for Python 3.8 and added support for Python 3.13.
Fix: Always check the `Requires-Python` metadata on PyPI or the official documentation for `qiskit-aer` and `qiskit` to ensure your Python environment is compatible.
gotcha Enabling GPU acceleration for `qiskit-aer` requires specific system configurations, including CUDA Toolkit and drivers. Setup can be complex and may require custom builds or specific installation flags.
Fix: Consult the official Qiskit Aer documentation for detailed GPU installation instructions specific to your operating system and hardware. Ensure all necessary NVIDIA CUDA components are correctly installed and configured.

Install

pip install qiskit-aer Install latest version

Imports

AerSimulator
```
from qiskit_aer import AerSimulator
```
Aer
```
from qiskit_aer import Aer
```
The `Aer` object is still used for getting specific simulator backends, but `AerSimulator` is the recommended direct import for the default simulator in newer Qiskit versions. The path `qiskit.providers.aer` is for legacy Qiskit versions.

Quickstart

This example demonstrates how to create a simple Bell state circuit, select the default AerSimulator backend, and execute the circuit to obtain measurement counts.

from qiskit import QuantumCircuit, transpile
from qiskit_aer import AerSimulator

# Construct quantum circuit
circ = QuantumCircuit(2, 2)
circ.h(0)
circ.cx(0, 1)
circ.measure([0,1], [0,1])

# Select the AerSimulator
simulator = AerSimulator()

# Run and get counts
compiled_circuit = transpile(circ, simulator)
job = simulator.run(compiled_circuit, shots=1000)
result = job.result()
counts = result.get_counts(circ)
print("Shot counts:", counts)

view raw JSON →