New York, USA · 9 November 2022 — IBM unveiled its newest quantum processor, codenamed Osprey, at the annual IBM Quantum Summit in New York. With 433 superconducting qubits, Osprey more than tripled the 127-qubit count of its predecessor, Eagle, which IBM had introduced in November 2021. The processor represented the largest superconducting quantum chip publicly disclosed at the time and marked a significant step in IBM's quantum hardware roadmap.
Alongside Osprey, IBM announced the IBM Quantum System Two — a new, modular quantum computing infrastructure designed to house multiple processors and enable networked quantum computing. The company also outlined its roadmap toward quantum-centric supercomputing, targeting systems with over 4,000 qubits by 2025.
Superconducting Qubit Architecture
Osprey uses superconducting transmon qubits — a type of qubit based on superconducting electrical circuits that operate at microwave frequencies. Like all superconducting quantum processors, Osprey must be cooled to extremely low temperatures to function. The chip operates at approximately 0.02 Kelvin (-273.13°C), just a fraction of a degree above absolute zero, to maintain the quantum coherence of its qubits.
This cooling requirement is achieved using a dilution refrigerator — a sophisticated cryogenic system that uses a mixture of helium-3 and helium-4 isotopes to reach temperatures colder than deep space. The refrigerator, which stands over three metres tall, houses the quantum chip at its base, surrounded by multiple stages of thermal shielding and extensive microwave control wiring.
The engineering challenge of scaling from 127 to 433 qubits is substantial. Each qubit requires dedicated control and readout lines, and the density of wiring, filters, and amplifiers within the cryostat increases dramatically with qubit count. IBM's engineers redesigned the cryogenic infrastructure, control electronics, and chip packaging to accommodate the larger processor while maintaining signal integrity and thermal isolation.
Beyond Qubit Count: The Quality Question
While the raw qubit count is the most visible metric, quantum computing researchers emphasise that qubit quality — measured by coherence time, gate fidelity, and error rates — is equally important, if not more so. A quantum computer's practical utility depends not merely on how many qubits it has, but on how reliably those qubits can perform operations and maintain quantum states.
IBM has been transparent about this distinction. The company publishes quantum processor performance metrics through its online platform, including median qubit coherence times (T1 and T2), gate error rates, and quantum volume scores — a holistic metric that captures both qubit count and quality.
IBM Quantum System Two
The Quantum Summit also marked the introduction of IBM Quantum System Two, a new platform designed to support the company's modular approach to quantum computing. Unlike the previous Quantum System One, which housed a single processor, System Two is built to accommodate multiple processors, advanced cryogenic infrastructure, and classical computing resources in an integrated architecture.
System Two's modular design reflects a fundamental shift in IBM's strategy. Rather than pursuing ever-larger single chips, the company envisions a future where multiple quantum processors are connected through quantum interconnects — communication links that transfer quantum information between chips. This networked approach, which IBM calls "quantum-centric supercomputing," aims to overcome the physical limitations of single-chip scaling by distributing computation across a system of interconnected processors.
The first IBM Quantum System Two unit became operational at IBM Research's Yorktown Heights facility in New York, housing Osprey and subsequent processors.
The Quantum Roadmap
IBM's quantum roadmap, first published in 2020 and updated at the 2022 summit, outlines a progression toward increasingly capable systems:
| Processor | Year | Qubits | Key Advance |
|---|---|---|---|
| Hummingbird | 2020 | 65 | Readout improvements |
| Eagle | 2021 | 127 | First processor over 100 qubits |
| Osprey | 2022 | 433 | 3.4× scaling, new cryogenics |
| Condor | 2023 | 1,121 | First processor over 1,000 qubits |
| Flamingo | 2025 | 1,386+ | Modular multi-chip connections |
The roadmap reflects IBM's belief that quantum computing will advance through a combination of processor scaling, modular interconnection, and error correction, rather than through any single breakthrough.
Quantum Computing and Africa
IBM's quantum computing programme has had a notable footprint in Africa. Through the IBM Quantum Network and the African Research Universities Alliance (ARUA), several African universities and research institutions have gained access to IBM's quantum computers via the cloud. Researchers in South Africa, for example, have used IBM quantum systems to study quantum chemistry, optimisation problems, and quantum machine learning.
The availability of cloud-based quantum computing has been particularly significant for African researchers, who might otherwise lack access to the multimillion-dollar hardware required for quantum experiments. IBM's Qiskit open-source framework, which allows users to programme quantum circuits in Python, has been widely adopted in university courses and research labs across the continent.
The Broader Quantum Landscape
IBM's Osprey announcement came amid intensifying global competition in quantum hardware. Google, which had announced quantum supremacy with its 53-qubit Sycamore processor in 2019, continued to develop its own superconducting qubit technology. Other major players included Rigetti Computing, IonQ (using trapped-ion qubits), Quantinuum (formed from the merger of Honeywell Quantum Solutions and Cambridge Quantum), and PsiQuantum (pursuing photonic quantum computing).
Each approach has distinct trade-offs. Superconducting qubits, used by IBM and Google, benefit from manufacturing compatibility with existing semiconductor fabrication techniques but face challenges with coherence times and error rates. Trapped-ion qubits, used by IonQ and Quantinuum, offer long coherence times and high gate fidelities but are harder to scale to large numbers. Photonic approaches promise room-temperature operation but face challenges in generating and detecting single photons reliably.
Sources
- IBM, "IBM Unveils 400 Qubit-Plus Quantum Processor and Next-Generation IBM Quantum System Two," 9 November 2022, ibm.com
- IBM Research Blog, "IBM Quantum Summit 2022," research.ibm.com
- IBM Quantum roadmap documentation, ibm.com/quantum
- Wikipedia, "IBM Quantum," en.wikipedia.org
- Post-Quantum, "IBM's Osprey Quantum Processor," postquantum.com
- ARUA-IBM Quantum Network partnership documentation, arua.org.au
FIRAT Editorial Board
Institutional Research Desk · Foresight Institute of Research and Translation
The collective editorial and research translation board of FIRAT, synthesising peer-reviewed evidence, policy briefs, and division milestones across our seven foundational research pillars.
