Amazon Braket now supports program sets, enabling quantum researchers to run complex workloads requiring hundreds of quantum circuit executions up to 24X faster. This new feature allows customers to submit up to 100 quantum programs or a single parametric circuit with up to 100 parameter values within a single quantum task. Program sets help minimize the time between subsequent circuit executions reducing quantum task processing overhead for complex algorithms.
Program sets are particularly valuable for researchers working with variational quantum algorithms (VQA), quantum machine learning models, and error mitigation techniques. Customers can create program sets using two approaches: submitting multiple independent circuits together, or submitting a single parametric circuit with parameter sets. Amazon Braket handles compilation and execution orchestration, returning results that include the status and outcomes for each quantum program. If individual programs within a program set fail during execution, customers receive partial results for successfully completed programs and error information for failed executions. When submitting a program set, you pay a single per-task fee accompanied by a per-shot fee based on the total number of successful shots across your workload in a program set.
Program sets are initially available on all superconducting quantum processing units (QPUs) from Rigetti in the US West (N. California) Region and IQM in the Europe (Stockholm) Region. Customers are able to submit program sets to Braket directly via the Amazon Braket SDK, from Qiskit via the Qiskit-Braket provider, or from PennyLane via the Amazon Braket PennyLane Plugin.
To learn more about program sets, visit the Amazon Braket developer guide, explore our new example notebooks, and visit our updated Amazon Braket management console.
Categories:
Source: Amazon Web Services
Latest Posts
- Amazon RDS for PostgreSQL supports minor versions 17.6, 16.10, 15.14, 14.19, and 13.22
- Noise suppression for dial-in participants in Teams audio conferences [MC1135397]
- Microsoft Teams | Manage voice and face recognition for rooms (MTR-W/MTRA) via device settings [MC1135396]
- Reporting labels retirement in Teams admin center [MC1135399]