IBM Unveils New Quantum Chips on the Path to Fault Tolerance
IBM introduced its 120-qubit Nighthawk processor and an experimental chip called Loon, and reported decoding quantum errors in real time, a key step toward fault-tolerant computing.
ScienceGNGV Editorial Team5 min readLast reviewed September 28, 2026
IBM announced a set of quantum computing advances on November 12, 2025, laying out concrete steps toward two long-promised goals: useful quantum advantage and, further out, fault-tolerant machines. At the center are two new processors. IBM Quantum Nighthawk packs 120 qubits linked by 218 next-generation tunable couplers, an arrangement the company says enables circuits with about 30 percent more complexity than its previous chips, supporting up to 5,000 two-qubit gates today and a planned 10,000 by 2027.
The second chip, IBM Quantum Loon, is experimental and aimed squarely at the hardest problem in the field: errors. Quantum bits are exquisitely fragile, and any practical quantum computer must constantly detect and correct mistakes. Loon is described as the first processor demonstrating all the key components needed for fault-tolerant quantum computing, including a new multi-layer architecture for routing connections between qubits.
Perhaps the most striking result is on the software side. IBM reported decoding quantum errors in real time, in under 480 nanoseconds, using a family of efficient codes known as qLDPC, and said it reached that milestone a full year ahead of its own schedule. Real-time decoding matters because corrections must keep pace with computations as they run; doing it fast enough is one of the central engineering hurdles between today’s noisy machines and reliable ones. IBM frames these results as building blocks toward its target of a fault-tolerant quantum computer by 2029.
As with all roadmaps, these are milestones rather than finished destinations, and a company’s announcements about its own products warrant the usual measured reading. Quantum computers still cannot outperform ordinary computers on most practical tasks. Even so, steady, verifiable progress on error correction, the discipline most experts agree is the true gatekeeper to useful quantum computing, is genuinely encouraging, and a sign that the field is maturing from physics experiment toward engineering reality.
