rewrite this title IBM Introduces Nighthawk and Loon Quantum Chips

IBM aims to reduce error rates in quantum computing, run complex circuits, and validate quantum supremacy by 2026 with its Nighthawk and Loon chips.

IBM has announced two new chips aimed at achieving error-free results in quantum computing: Nighthawk and Loon. The company expects these chips, which operate with different quantum bit (qubit) connection methods, to produce fewer errors and support more complex calculations. According to IBM, these developments could be an indicator that “quantum supremacy” can be achieved by the end of 2026—meaning a quantum computer can solve problems faster than a classical computer.

IBM’s Move with Nighthawk and Loon

Among the new chips, IBM Quantum Nighthawk stands out as the chip IBM believes can demonstrate quantum supremacy. IBM states that the version of Nighthawk it will offer to partners at the end of 2025 will connect 120 qubits with 218 next-generation adjustable couplers. These couplers link qubits with their neighbors in a square grid structure.

IBM says that thanks to this structure, Nighthawk can run 30% more complex circuits while keeping error rates low and can solve computation problems requiring up to 5,000 two-qubit gates. It will be developed to support 7,500 two-qubit gates in 2026, 10,000 in 2027, and 15,000 in 2028. Nighthawk will also enable the connection of more than 1,000 qubits using long-range couplers.

The other chip, IBM Quantum Loon, presents a more experimental approach. Loon can connect qubits not only horizontally but also vertically. According to IBM, regardless of which chip proves more useful, the additional connection options mean fewer errors and more complex calculations. This could bring quantum computers closer to real-world applications.

IBM also announced that, along with the new chips, it will contribute to a community-driven quantum supremacy tracking platform, supported by researchers from Algorithmiq, Flatiron Institute, and BlueQubit. The company states that this platform supports three different quantum supremacy experiments: observable estimation, variational problems, and efficient classical verification problems.

IBM’s quantum software, Qiskit, provides developers with more control than ever before. Dynamic circuits achieve 24% higher accuracy, and the cost of extracting accurate results is reduced 100-fold with HPC-powered error mitigation.

Different Approaches from Competitors

On the other hand, IBM’s approach differs from competitors like Google. IBM focuses on connecting qubits in small groups. Google, however, had announced in 2024 that it could theoretically demonstrate quantum supremacy with its Willow chip. In 2025, Google announced the Quantum Echoes algorithm for Willow, declaring that it had run the “first verifiable quantum supremacy” using the OTOC (out-of-order time correlator) algorithm.

What is a Quantum Chip?

A quantum chip (or quantum processing unit, QPU) is the “brain” or core component of a quantum computer.

Unlike a classical computer chip that uses “bits” (which are either a 0 or a 1), a quantum chip uses “qubits” (quantum bits).

Because of quantum mechanics, qubits can exist in multiple states at once (a 0 and a 1 at the same time, known as superposition) and can be linked together (known as entanglement). These properties allow the chip to perform many calculations simultaneously, giving it the potential to solve extremely complex problems—like drug discovery or advanced materials science—that are impossible for even the most powerful classical supercomputers.

You Might Also Like;