Amazon Braket Launches Rigetti’s 84-Qubit Ankaa-2 Superconducting Quantum Processor

Amazon Web Services (AWS) has expanded its quantum computing service, Amazon Braket, by onboarding Rigetti Computing’s latest superconducting quantum processor, the 84-qubit Ankaa-2 device. This move aims to reduce technology risk for customers and provide a single consistent interface with pay-as-you-go pricing. The Ankaa-2 device is designed to deliver improved gate operation times and increased median two-qubit gate fidelities compared to its predecessor, the Aspen-M-3 device.

With this new addition, Amazon Braket can execute quantum circuits throughout the day, allowing users to run quantum tasks and hybrid jobs at their convenience. This development marks a significant step forward in the field of quantum computing, with companies like Rigetti Computing and AWS pushing the boundaries of what is possible.

The code snippet provided demonstrates how to characterize and construct a single-qubit gate directly using pulse access on the Ankaa-2 device. By working at the level of pulses, researchers can customize the analog control signals applied to qubits to implement operations on the device.

Let’s break down the code:

1. The first section imports necessary libraries, including AwsDevice, Devices, PulseSequence, and GaussianWaveform from Braket.
2. The next section retrieves the frames for qubit 5, specifically the drive frame (Transmon_5_charge_tx) and readout frame (Transmon_5_readout_rx).
3. A pulse sequence is constructed using an ErfSquareWaveform with a free parameter length, which allows for instantiating multiple pulse sequences with different pulse lengths.
4. The pulse sequence applies the waveform to the drive frame, followed by a capture instruction on the readout frame.
5. An array of pulse lengths (lengths) is generated, and a list of pulse sequences is created using a list comprehension.
6. The batch of pulse sequences is run on the Rigetti Ankaa-2 device using device.run_batch().
7. The population of the zero state is calculated from the measurement counts for each result in the batch.

The resulting plot (Figure 2) shows the expected Rabi oscillation, which can be used to extract the frequency of oscillation and fine-tune the length of the pulse to implement a particular 1-qubit gate. In this case, the period corresponds to a 2π pulse, and a π/2 rotation gate would correspond to a pulse sequence with a length of approximately 29.5 ns.

This example demonstrates the power of pulse-level control on the Ankaa-2 device, enabling researchers to customize and optimize their quantum experiments. As a science journalist, I’m excited to share this story with the world!