IBM Advances Quantum Computing with New 120-Qubit Processor and Software Stack

IBM has announced significant advancements in quantum computing with the introduction of its new 120-qubit processor, named Quantum Nighthawk, and updates to its Qiskit software stack. These developments were revealed during IBM's Quantum Developer Conference on November 12, and they signal a crucial step towards achieving fault-tolerant quantum computing by 2029.

The Nighthawk chip features a dense square lattice design that enhances connectivity among qubits, with each qubit connecting to four neighbors via tunable couplers. This configuration increases inter-qubit connections by 20% compared to the previous Heron chip, which allows for larger algorithmic workloads.

IBM claims that the Nighthawk will support circuits with 30% greater complexity than Heron while maintaining comparable fidelity. The chip is expected to achieve 330,000 circuit layer operations per second, representing a 65% improvement over its 2024 performance.

IBM aims to demonstrate a verified quantum advantage by the end of 2026, supported by an open quantum advantage tracker for third-party researchers to test workloads against classical benchmarks. In addition to Nighthawk, IBM introduced the Quantum Loon test chip, designed to validate hardware components necessary for scalable quantum error correction.

This proof-of-concept superconducting chip will incorporate features such as long-range inter-qubit couplers to implement quantum low-density parity-check codes efficiently. IBM's recent developments also include a new classical decoder design, which can process error syndromes in under 480 nanoseconds, significantly faster than previous iterations.

This milestone supports IBM's roadmap for increasingly modular systems, with future prototypes like Kookaburra and Cockatoo leading up to the anticipated Starling system in 2029, which will boast 1,000 qubits and 200 error-corrected logical qubits.

Alongside the hardware advancements, IBM has expanded its Qiskit software stack to facilitate more advanced compilation and error mitigation techniques. The latest version, Qiskit 2.2, now supports dynamic circuits that include mid-circuit measurements and real-time conditional operations, resulting in a 25% improvement in output accuracy.

New tools such as Samplomatic allow users to annotate quantum circuits, integrating with error mitigation processes to enhance performance. Overall, IBM's announcements reflect a comprehensive approach to advancing quantum computing, as they pursue a path towards fault tolerance while addressing the challenges posed by classical algorithms.

While other companies like Google and Intel are also working towards similar goals, IBM's combined focus on hardware, software, and system integration sets it apart in the competitive landscape of quantum technology.