IBM unveiled its path to build the world's first large-scale, fault-tolerant quantum computer, setting the stage for practical and scalable quantum computing. Delivered by 2029, IBM Quantum Starling will be built in a new IBM Quantum Data Center in Poughkeepsie, New York and is expected to perform 20,000 times more operations than today's quantum computers. To represent the computational state of an IBM Starling would require the memory of more than a quindecillion (1048) of the world's most powerful supercomputers.
With Starling, users will be able to fully explore the complexity of its quantum states, which are beyond the limited properties able to be accessed by current quantum computers. IBM, which already operates a large, global fleet of quantum computers, is releasing a new Quantum Roadmap that outlines its plans to build out a practical, fault-tolerant Quantum computer. A large-scale, fault-Tolerant quantum computer with hundreds or thousands of logical qubits could run hundreds of millions to billions of operations, which could accelerate time and cost efficiencies in fields such as drug development, materials discovery, chemistry, and optimization.
Starling will be able to access the computational power required for these problems by running 100 million quantum operations using 200 logical qubits. Able to prepare and measure logical qubits through computation. Capable of applying universal instructions to these logical qubits.
Able to decode measurements from logical qubits in real-time and can alter subsequent instructions. Modular to scale to hundreds or thousands of logical qubit to run more complex algorithms. Efficient enough to execute meaningful algorithms with realistic physical resources, such as energy and infrastructure. IBM is introducing two new technical papers that detail how it will solve the above criteria to build a large-scale, fault- tolerant architecture.
The first paper unveils how such a system will process instructions and run operations effectively with qLDPC codes. This work builds on a groundbreaking approach to error correction featured on the cover of Nature that introduced quantum low-density parity check (qLDPC) codes. This code drastically reduces the number of physical qubits needed for error correction and cuts required overhead by approximately 90%, compared to other leading codes.
Additionally, it lays out the resources required to reliably run large-scale quantum programs to prove the efficiency of such an architecture over others. The second paper describes how to efficiently decode the information from the physical qubits and charts a path to identify and correct errors in real-time with conventional computing resources. The new IBM Quantum Roadmap outlines the key technology milestones that will demonstrate and execute the criteria for fault tolerance.
Each new processor in the roadmap addresses specific challenges to build quantum computers that are modular, scalable, and error-corrected: IBM Quantum Loon, expected in 2025, is designed to test architecture components for the qLDPC code, including "C-couplers" that connect qubits over longer distances within the same chip. IBM Quantum Kookaburra, expected in 2026, will be IBM's first modular processor designed to store and process encoded information. It will combine quantum and a single processor designed to store and process encrypted information.
It will combine quantum data from the physical qubits and data from the QLDPC code.
