Skip to main content

Amazon Braket

What is it​

Amazon Braket is the quantum computing service from AWS, with an open-source Python SDK (amazon-braket-sdk). Its goal is portability through the cloud: from one program you can target a fast local simulator, AWS-managed on-demand simulators, or real quantum hardware from several vendors — all billed and accessed through your AWS account.

Braket circuits are written in a vendor-neutral way, and the same code can be retargeted to different devices by swapping the device object.

Install​

pip install amazon-braket-sdk

The local simulator runs without any AWS account. To use managed simulators or hardware you need AWS credentials configured and the relevant device enabled in your account/region.

Hello Quantum​

A Bell state 12(∣00⟩+∣11⟩)\frac{1}{\sqrt{2}}\left( \lvert 00 \rangle + \lvert 11 \rangle \right) on the LocalSimulator:

from braket.circuits import Circuit
from braket.devices import LocalSimulator

# Build the Bell circuit
bell = Circuit().h(0).cnot(control=0, target=1)
print(bell)

# Run locally — no AWS account required
device = LocalSimulator()
result = device.run(bell, shots=1024).result()

print(result.measurement_counts)
# e.g. Counter({'00': 511, '11': 513})

To run the same circuit on a managed device, you swap the device for an AwsDevice ARN, for example an on-demand simulator (SV1) or a hardware device:

from braket.aws import AwsDevice

# On-demand managed state-vector simulator (incurs AWS charges)
sv1 = AwsDevice("arn:aws:braket:::device/quantum-simulator/amazon/sv1")
task = sv1.run(bell, shots=1024)
print(task.result().measurement_counts)

Strengths & use cases​

  • Multi-vendor access. A single SDK reaches gate-model and analog devices from vendors such as IonQ, IQM, Rigetti, and the neutral-atom analog device from QuEra.
  • Managed on-demand simulators. AWS hosts SV1 (state vector), DM1 (density matrix), and TN1 (tensor network) for circuits too large for a laptop.
  • Local development. The LocalSimulator lets you prototype offline before spending on cloud resources.
  • AWS integration. Hybrid Jobs, S3 result storage, and IAM-based access fit existing AWS workflows.

Ecosystem​

  • Braket SDK — circuit construction and device execution.
  • Amazon Braket Hybrid Jobs — managed execution of variational/hybrid algorithms.
  • PennyLane plugin (pennylane-braket) — run differentiable circuits on Braket devices.
  • Amazon Braket Algorithm Library — reference implementations of common algorithms.

Resources​

Try a cloud-backed exercise in the labs, or plan your path in the roadmaps.