IBM unveiled its most powerful quantum computing system to date, Osprey, a 433-qubit machine that supports three times as many qubits as the current Eagle system, and reiterated its plans to launch a 1,121-qubit machine called Condor. Qubit processors in 2023.
At the 2022 IBM Quantum Summit, the company also said it was continuing to develop a modular quantum platform called System Two, which combines multiple processors into a system with new communication links. IBM said it will use a hybrid cloud Middleware to integrate quantum and classical workflows.
Additionally, IBM said it will continue to prototype quantum software applications for specific use cases. IBM says that by 2025, developers will be able to explore quantum machine learning applications and more.
Big Blue is following a quantum roadmap laid out earlier this year that sets long-term goals. It includes developing a 4,000-qubit system by 2025 built from a cluster of quantum processors and working with partners to develop software that can control quantum systems and network them together while eliminating errors that can slow down quantum efforts.
“We see three main application areas for quantum computers today: simulating nature, such as materials and chemistry; solving problems around data structures, such as some in optimization and machine learning; and problems in pure mathematics and number theory,” IBM said Oliver Dial, chief hardware architect at Quantum. “This is where we see immediate use cases. But the more people who use quantum, the more use cases will be found.”
While advances in hardware have driven the development of quantum needles, software will play an important role in advancing the technology.
This week, IBM also released a beta update for Qiskit Runtime, which IBM calls “a quantum computing service and programming model that allows users to optimize workloads and efficiently execute them on large-scale quantum systems.” With the update, it Now includes a feature that allows users to trade speed with a simple change to the API to reduce error counts.
“Quantum computers have higher error rates than classical computers because of the subtle nature of quantum states,” Dial said. “We typically talk about gate fidelity—about the percentage of time a quantum computer gives the correct answer after one operation. Today’s state-of-the-art is about 99.9 percent, or one error per thousand gates,” Dial said.
“Techniques like error mitigation allow us to increase the effective fidelity of the device by running the circuit multiple times and combining answers, allowing us to trade off quality and scale/speed. For the user, this makes Quantum is more flexible to use. They can decide which combination of mass and scale/speed works best for them,” Dial said.
The goal is complete quantum error correction. At the same time, these technologies allow users to derive real value from quantum machines in the short term, Dial said.
Industry partners will also play an important role in developing quantum systems. This week, IBM said it has partnered with Vodafone to develop quantum use cases for telecommunications. The telecommunications provider will also deploy IBM Quantum Safe cryptography technology in its systems.
Vodaphone will become part of the IBM Quantum Network, which includes 200 organizations and more than 450,000 users. In addition to Vodafone, the network has added French bank Crédit Mutuel to explore use cases in financial services, and Swiss R&D agency uptown Basel to promote the development of quantum skills.