A hardware-agnostic quantum computing framework, inspired by the architecture of NVIDIA's CUDA-Q, designed to provide a unified interface for executing quantum circuits across a wide range of third-party quantum processing units (QPUs).

The core goal of rocQuantum is to offer a seamless user experience where a quantum circuit can be defined once and then executed on different hardware backends—from remote cloud platforms to local simulators—with minimal changes to the code. This is achieved through a robust backend abstraction layer that supports multiple integration models:

• Type A (Remote API): For API-based services (e.g., IonQ, Pasqal), rocQuantum manages the HTTP client, authentication, and job lifecycle internally.
• Type B (Local SDK): For providers that require a local SDK (e.g., Quantum Brilliance's Qristal), rocQuantum interfaces directly with the provider's tools for synchronous execution.
• Type C (Cloud Intermediary): For platforms accessed via a cloud provider's SDK (e.g., Rigetti on AWS Braket), rocQuantum handles the interaction with the intermediary service.

1. Clone the repository:

   git clone <repository-url>
   cd rocQuantum-1

2. Install the required Python packages:

   pip install requests boto3

   (Note: Specific provider SDKs like qiskit or cirq should be installed as needed).

To access third-party backends, you must configure your credentials using environment variables.

• For IonQ:

  export IONQ_API_KEY="YOUR_IONQ_API_KEY"

• For Quantinuum:

  export CUDAQ_QUANTINUUM_CREDENTIALS="YOUR_USERNAME,YOUR_PASSWORD"

• For Rigetti (via AWS Braket): The Rigetti backend uses the boto3 library, which automatically finds AWS credentials. Configure them using standard AWS methods, such as:

  • Environment Variables:

    export AWS_ACCESS_KEY_ID="YOUR_AWS_ACCESS_KEY"
    export AWS_SECRET_ACCESS_KEY="YOUR_AWS_SECRET_KEY"
    export AWS_SESSION_TOKEN="YOUR_SESSION_TOKEN" # (Optional)

  • Shared Credentials File (~/.aws/credentials)

• For Pasqal & Infleqtion: Set PASQAL_API_KEY and SUPERSTAQ_API_KEY respectively.

A simple Command-Line Interface is provided for quick demonstrations. It runs a standard Bell State circuit on the specified backend.

Example:

python rocq_cli.py run --backend ionq --shots 100

This command will:

1. Check for the IONQ_API_KEY.
2. Target the IonQ simulator.
3. Submit a Bell State circuit for 100 shots.
4. Poll for and print the final measurement results.

The framework is designed for extensibility. The following backends are currently integrated:

The script in examples/run_bell_state.py demonstrates the power of the framework by defining a single circuit and running it on three different backend types: IonQ (API), Qristal (local SDK), and Rigetti (cloud intermediary).