To get started with quantum computing on AWS, enable Amazon Braket, choose a managed Jupyter notebook or a local Python setup, and run a small circuit on a simulator before trying a QPU. A Bell-state circuit is a useful first exercise: it produces correlated measurement results, and Braket saves task results to an S3 bucket in your AWS account.
How do I get started with Amazon Braket?
Amazon Braket is an AWS service for submitting quantum-computing work to simulators and quantum processing units (QPUs). Its basic unit is a quantum task. For a gate-based circuit, a task includes the circuit, measurement instructions, the number of shots, and request metadata. For analog Hamiltonian simulation, it describes a register layout and time- and space-dependent control fields.
You can define, submit, and monitor tasks in a notebook with the Amazon Braket SDK, or use the AWS console. The SDK provides a convenient layer over the Braket API and Boto3. When a device processes a task, its results are stored in an S3 bucket in your AWS account. See the Amazon Braket overview.
Choose where to write and run code
- Managed notebook: Braket notebooks are Jupyter environments based on SageMaker AI notebook instances. Notebooks created through the console come with the SDK and dependencies preloaded. The notebook instance itself can incur AWS charges.
- Local Python environment: Install the
amazon-braket-sdkpackage with pip, then configure access to AWS. AWS also documents a PennyLane plugin package for workflows that use PennyLane.
Follow AWS’s setup instructions to enable the service and start with either environment. A managed notebook is optional; it is not required to submit Braket tasks.
Free tools Windows power users keep installed
One-click scans. No signup required.
#1 Best Overall
How do I run my first quantum circuit on AWS?
Start with the official Building your first circuit example, which creates a Bell state. This small circuit demonstrates how quantum operations are assembled and how measurement counts are returned.
- Create a circuit. Import the Braket SDK and define a circuit that applies a Hadamard gate to the first qubit, then a controlled-NOT gate from the first qubit to the second.
- Select a simulator. For a first run, use a local simulator to check that the circuit and result-handling code work. You can also submit to an on-demand simulator such as SV1 when you need managed execution.
- Set the shots and submit. Shots are repeated circuit executions used to gather measurement outcomes. The SDK workflow is to choose a device, instantiate the circuit, call its run method, and collect the result.
- Inspect the counts. The Bell-state example yields outcomes concentrated on
00and11, typically in roughly equal proportions. Finite shots introduce variation, so an exactly even split is not guaranteed.
A simulator is the sensible first stop: AWS recommends checking work on simulators before QPU use to catch coding or configuration errors without incurring QPU task charges. Simulator execution is not necessarily free, and notebook compute, S3 storage, and other AWS services may also add charges.
Rank #2
Can I try quantum computing on a simulator before using a real quantum computer?
Yes. Braket offers local and on-demand simulators, as well as QPUs and embedded simulators. Pick according to what you are trying to learn: local execution for rapid prototyping, managed simulation for larger or different simulation needs, or a QPU when you specifically need to experiment with physical quantum hardware.
| Option | Useful for | AWS-documented capability | What to keep in mind |
|---|---|---|---|
| Local state-vector simulator | Rapid prototyping and debugging on your own computer | Up to 25 qubits, depending on host hardware | Actual capacity and speed depend on the machine and circuit. |
| SV1 on-demand state-vector simulator | Managed state-vector simulation | Up to 34 qubits; AWS says a dense 34-qubit circuit of depth 34 may take around one to two hours, depending on gates and other factors | This is an AWS capability and timing estimate, not a performance guarantee for every program. |
| DM1 on-demand density-matrix simulator | Density-matrix simulation | Up to 17 qubits | Choose it when the simulation method fits your task, rather than treating qubit capacity alone as the deciding factor. |
| QPU | Experiments on physical quantum hardware | Varies by device; check current device properties | Supported operations, technology, Region, availability window, and queue can affect whether and when a task can run. |
These simulator figures are capabilities listed in the AWS simulator guide, not a promise that a particular circuit will run at that scale on your hardware or within a particular time.
The Tool Desk
Outbyte PC Repair FREEClear out junk files and repair common Windows errorsFree Scan →Outbyte Driver Updater FREEScan for outdated or missing drivers - takes under a minuteDriver Scan →How do I choose a Braket device?
Choose a simulator or QPU based on the circuit and the purpose of the run—not on the assumption that a QPU is automatically the next step after a simulator. Before submission, check the selected device’s current properties and availability.
- Purpose: Use a local or on-demand simulator to debug or explore simulation. Choose a QPU only when physical-hardware behavior is part of the experiment.
- Operations and result types: Confirm that the device supports the gates and result types your task requires.
- Capacity and method: Match circuit size and the simulator method to the option’s documented capabilities.
- Technology and availability: The developer guide identifies QPU providers including AQT, IonQ, IQM, QuEra, and Rigetti. Device inventory and availability windows can change, and a hardware task may wait for a device window.
- Region: A device may be in a different Region from your working environment. The SDK can submit to a QPU in another Region by creating a session for the device’s Region.
Device status is a snapshot of current conditions, not a permanent attribute. Use the device guide and the console’s current device details to confirm what is available before you run.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.What can Amazon Braket cost?
Braket has no upfront commitment for device access; charges are based on usage. Budget for more than the quantum task itself: managed notebook compute, simulator use, storage, and other AWS resources can contribute to the bill.
AWS provides near-real-time cost tracking estimates and optional per-device spending limits for QPU tasks. Those limits do not cover simulator tasks, managed notebooks, Hybrid Job EC2 instance costs, or Braket Direct reservations. Estimates can differ from actual charges and do not include every discount, credit, or cost from other AWS services. Check the current Amazon Braket pricing and cost-management information before running billable work.
Do these 3 things before closing this tab:
1Fix the driver behind crashes, sound loss and screen glitches2Repair Windows errors before they cause bigger problems3Scan for outdated or missing drivers - takes under a minuteBest Value
Set safeguards before experimenting
- Use simulators to verify code before submitting QPU tasks.
- Use AWS IAM to control who can access devices.
- Set AWS Budgets alerts to help monitor account spending.
- When reviewing quantum-task costs in the console, check each relevant Region: the console shows tasks only for the currently selected Region.
These safeguards complement, rather than replace, checking notebook, simulator, storage, and other AWS charges.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
