> ## Documentation Index
> Fetch the complete documentation index at: https://docs.qbraid.com/llms.txt
> Use this file to discover all available pages before exploring further.
# BraketProvider
> Runtime integration for streamlined access to Amazon Braket supported devices.
API Reference:
[qbraid.runtime.aws](https://qbraid.github.io/qBraid/stubs/qbraid.runtime.aws.html)
## Installation & Setup
To interface with Amazon Braket [supported devices](https://docs.aws.amazon.com/braket/latest/developerguide/braket-devices.html),
install the `braket` extra:
```bash theme={null}
pip install 'qbraid[braket]'
```
Then, follow [instructions](https://github.com/aws/amazon-braket-sdk-python#boto3-and-setting-up-aws-credentials)
to configure your AWS credentials.
## Basic Usage
Submit a Quantum Task to an AWS device using the `BraketProvider`:
```python theme={null}
from qbraid.runtime import BraketProvider
provider = BraketProvider()
provider.get_devices()
# [,
# ,
# ,
# ,
# ,
# ,
# ,
# ,
# ,
# ]
device = provider.get_device("arn:aws:braket:::device/quantum-simulator/amazon/sv1")
type(device)
# qbraid.runtime.aws.device.BraketDevice
device.metadata()
# {'device_id': 'arn:aws:braket:::device/quantum-simulator/amazon/sv1',
# 'device_type': 'SIMULATOR',
# 'num_qubits': 34,
# 'provider_name': 'Amazon Braket',
# 'status': 'ONLINE',
# 'queue_depth': 0}
```
Now that we've instantiated our device, in this case the AWS SV1 simulator, we can construct a quantum circuit and submit a task using the `.run` method:
```python theme={null}
from braket.circuits import Circuit
circuit = Circuit().h(0).cnot(0, 1)
job = device.run(circuit, shots=10)
```
We now have `job` which is of type `BraketQuantumTask`, which inherits from `QuantumJob`. To see the results,
we can do the following:
```python theme={null}
res = job.result()
res.data.get_counts()
# {'00': 6, '11': 4}
res.data.measurements
# array([[0, 0],
# [0, 0],
# [1, 1],
# [1, 1],
# [0, 0],
# [0, 0],
# [0, 0],
# [1, 1],
# [0, 0],
# [0, 0]])
```
See how to visualize these results in the [Visualization](/v2/sdk/user-guide/visualization#plot-experimental-results) section.
## Runtime Options
The `BraketDevice.run()` method accepts additional keyword arguments that are passed through
to the underlying Amazon Braket SDK's `AwsDevice.run()`. These options let you control
device-specific behavior such as qubit rewiring, error mitigation, experimental capabilities, and more.
### Disable Qubit Rewiring
By default, Braket may remap the logical qubits in your circuit to physical qubits on the device.
To force the use of the exact qubits specified in your circuit, disable qubit rewiring:
```python theme={null}
job = device.run(circuit, shots=1000, disable_qubit_rewiring=True)
```
This is required when using [verbatim compilation](https://docs.aws.amazon.com/braket/latest/developerguide/braket-openqasm-verbatim-compilation.html)
or pulse-level gate definitions.
### Error Mitigation (IonQ)
IonQ devices on Braket support error mitigation through debiasing, which creates
variations of your circuit to reduce the effects of hardware noise:
```python theme={null}
from braket.error_mitigation import Debias
device = provider.get_device("arn:aws:braket:us-east-1::device/qpu/ionq/Aria-1")
job = device.run(
circuit,
shots=2500, # Minimum 2500 shots required for debiasing
device_parameters={"errorMitigation": Debias()},
)
```
### Experimental Capabilities
Some devices offer experimental features that must be explicitly enabled. For example,
QuEra Aquila supports tight atom geometries and local detuning, and IQM devices support
dynamic circuits with mid-circuit measurement:
```python theme={null}
device = provider.get_device("arn:aws:braket:us-east-1::device/qpu/quera/Aquila")
job = device.run(
program,
shots=1000,
experimental_capabilities="ALL",
)
```
See [Experimental Capabilities](https://docs.aws.amazon.com/braket/latest/developerguide/braket-experimental-capabilities.html) for a full list of available features per device.
### Parametric Circuits
For OpenQASM programs with input parameters, pass values at runtime:
```python theme={null}
job = device.run(
openqasm_program,
shots=1000,
inputs={"theta": 1.57, "phi": 0.785},
)
```
### Reservation ARN
If you have a [Braket Direct](https://aws.amazon.com/braket/direct/) reservation for
exclusive device access, provide the reservation ARN:
```python theme={null}
job = device.run(
circuit,
shots=1000,
reservation_arn="arn:aws:braket:us-east-1:123456789012:reservation/abc123",
)
```
### Tags
Attach metadata tags to your quantum tasks for tracking and organization:
```python theme={null}
job = device.run(
circuit,
shots=1000,
tags={"experiment": "bell-state", "team": "quantum-research"},
)
```
### Options Reference
| Parameter | Type | Description |
| --------------------------- | ------------------ | ---------------------------------------------- |
| `disable_qubit_rewiring` | `bool` | Force use of exact qubits without remapping |
| `device_parameters` | `dict` | Device-specific config (e.g. error mitigation) |
| `experimental_capabilities` | `str` | Set to `"ALL"` to enable experimental features |
| `inputs` | `dict[str, float]` | Parameter values for parametric circuits |
| `reservation_arn` | `str` | Braket Direct reservation ARN |
| `tags` | `dict[str, str]` | Metadata tags for the quantum task |
### Via qBraid Runtime API
When submitting jobs through the [QbraidProvider](/v2/sdk/user-guide/providers/native), these same
options can be passed via the `runtimeOptions` field. The runtime API forwards them as keyword
arguments to the Braket `device.run()` call:
```python theme={null}
from qbraid.runtime import QbraidProvider
provider = QbraidProvider()
device = provider.get_device("aws:quera:qpu:aquila")
job = device.run(
program,
shots=1000,
runtime_options={"experimental_capabilities": "ALL"},
)
```
```python theme={null}
device = provider.get_device("aws:ionq:qpu:aria-1")
job = device.run(
circuit,
shots=2500,
runtime_options={"device_parameters": {"errorMitigation": "Debias"}},
)
```