Overview
Adding to the available hardware platforms, IonQ Cloud is now accessible via Haiqu SDK to use IonQ’s trapped-ion QPUs (Quantum Processing Units). This integration allows users to leverage the computational power of IonQ’s quantum processors directly from the Haiqu SDK environment.How to Access QPUs via IonQ Cloud
The integration is designed to be user-friendly, with a focus on simplifying the process of accessing quantum resources. Below is a step-by-step guide to using Haiqu SDK for this purpose.Obtain an IonQ Cloud API key
Users must acquire an access token from IonQ. Should you want to run your workloads via IonQ Cloud, make sure to create an API key on IonQ Cloud here. This token serves as a secure credential to authorize access to the quantum processors.
Configure Haiqu SDK
To connect Haiqu SDK with IonQ Cloud QPUs, users need to input the obtained access token into the options as
ionq_api_key. The token can be passed as a variable directly through the SDK’s API.Submit Quantum Jobs
Once authenticated, users can submit quantum circuits or algorithms to IonQ QPUs directly through Haiqu SDK. The SDK handles all backend operations, including job queuing, execution, and result retrieval.
Make sure to check which devices are available in your environment by calling
haiqu.list_devices() first!Using the IonQ Simulator
Alongside the physical QPUs, IonQ Cloud exposes a cloud simulator under the device IDionq_simulator. It runs on IonQ Cloud, so it uses the same ionq_api_key credential as the QPUs (it is not one of the credential-free local simulators).
By default the simulator is ideal (noiseless). To model the noise of a specific IonQ device, pass a noise_model string identifier — the name of a characterized IonQ device (e.g. "aria-1") — inside options.
The
noise_model for ionq_simulator must be a string identifier naming an IonQ device. (This differs from aer_simulator, which expects a dict or a qiskit_aer.NoiseModel object.) Noisy simulation is capped at 12 qubits, enforced server-side.