In brief
- IBM introduced Nighthawk and Loon quantum processors on Wednesday.
- Nighthawk’s 120 qubits and 218 couplers support circuits up to 5,000 two-qubit gates.
- IBM aims for community-verified quantum advantage by 2026 and milestones toward fault tolerance by 2029.
IBM debuted the next steps on its roadmap for practical quantum computing on Wednesday, unveiling upgraded processors, software, and fabrication methods that it said will help push the field toward a verified quantum advantage by 2026 and milestones on the way to fault tolerance by 2029.
“Quantum advantage” refers to the point at which a quantum computer performs a task that no traditional computer can match. Fault tolerance is the ability of a quantum computer to keep its performance stable in the face of errors. If IBM’s roadmap holds, then IBM’s Nighthawk processor would mark a crucial step towards a commercially viable quantum computer by the end of the decade.
While IBM’s announcement puts quantum computing a step closer to “Q-Day,” the new processors are still far from a threat to the encryption protecting Bitcoin.
Cracking Bitcoin’s elliptic curve cryptography would require a fault-tolerant quantum computer with roughly 2,000 logical qubits, which is equivalent to tens of millions of physical qubits once error correction is factored in. The Quantum Nighthawk is a 120-qubit processor designed to handle more complex computations while maintaining low error rates.
Still, Q-Day is coming closer. The first Nighthawk systems are expected to reach users by the end of 2025, with future iterations projected to exceed 1,000 connected qubits by 2028. The chip connects each qubit through 218 tunable couplers, about 20% more than IBM’s previous Heron design in 2023. IBM said the new architecture allows circuits roughly 30% more complex, supporting computations of up to 5,000 two-qubit gates.
The Nighthawk is the next waypoint in IBM’s Starling roadmap, a series of steps announced in July to deliver a large-scale, fault-tolerant quantum computer—IBM Quantum Starling—by 2029. Reaching the goal of manufacturing a scalable quantum computer for industrial use requires significant advancements in modular architecture and error correction, among other advancements anticipated in the Starling build-out.
IBM’s announcement followed a wave of renewed investment in quantum computing. In October, Google said its Willow processor achieved a verified quantum speed-up, completing a physics simulation faster than any known classical supercomputer. This result renewed fears over the long-term security of Bitcoin’s encryption.
To support its quantum ambitions, IBM partnered with Algorithmiq, the Flatiron Institute, and BlueQubit to launch a quantum-advantage tracker, an open-source platform for comparing quantum and classical results across benchmark experiments.
IBM also announced that it is expanding its Qiskit software to match the new hardware. The company said dynamic circuits in Qiskit improved accuracy by 24% at the 100-qubit scale. A new C-API interface links Qiskit with high-performance classical systems to accelerate error mitigation, reducing the cost of extracting accurate results, IBM claims, by more than 100 times.
By 2027, IBM plans to add computational libraries for machine learning and optimization to help researchers model physical and chemical systems.
Building towards fault tolerance
IBM also announced progress on its experimental Quantum Loon processor, which the company said demonstrates all key hardware components needed for fault-tolerant quantum computing. The chip architecture builds on technologies already proven in other test systems, including long-range “c-couplers” that link distant qubits and the ability to reset qubits between operations.
The company reported a tenfold speedup in error-decoding performance, achieving real-time correction under 480 nanoseconds using qLDPC codes—a milestone it said came a year ahead of schedule.
To accelerate development, IBM moved production of its quantum chips to a 300-millimeter wafer line at the Albany NanoTech Complex in New York. The transition, it said, has doubled research speed, increased chip complexity tenfold, and enabled multiple processor designs to be developed and explored in parallel.
IBM said the updates mark continued progress toward scalable, fault-tolerant quantum systems and provide the groundwork for community-verified demonstrations of quantum advantage in the next several years.
“We believe that IBM is the only company that is positioned to rapidly invent and scale quantum software, hardware, fabrication, and error correction to unlock transformative applications,” IBM Research Director Jay Gambetta said in a statement.
Generally Intelligent Newsletter
A weekly AI journey narrated by Gen, a generative AI model.
