IBM publishes its quantum numbers: 90-plus machines, 240,000 users and a claim for the end of 2026
The company has set out how many systems and users it has worldwide. The substance is in the timeline: IBM expects quantum advantage to be verified by the scientific community by the end of this year.
IBM has published figures for its quantum computing platform: more than 90 quantum computers deployed worldwide, over 240,000 users, more than 325 ecosystem partners, and roughly 4 trillion quantum circuit executions to date. These are the company's own figures and there is no independent verification.
The scale of the numbers can mislead, so one note: executing a quantum circuit is a small operation lasting seconds. The 4 trillion figure shows how much the platform is used, not the size of the problems being solved.
What separates quantum from classical
A classical computer holds information as 0s and 1s and is in one state at each step. A quantum computer uses qubits, which can carry several states at once until they are measured. This does not make quantum faster at everything; it will not open your email sooner. The advantage applies to particular classes of problem: optimisation, cryptography and molecular simulation. In exchange, qubits are fragile; the system has to be held near absolute zero and isolated from vibration and magnetic fields.
Where the technology actually stands
IBM's latest hardware, Nighthawk, carries 120 qubits and 20 percent more tunable couplers than the previous generation, which the company says allows circuits 30 percent more complex. A second processor, Loon, is not built for computation but as a testbed for error correction, trialling a quantum error-correcting scheme known as qLDPC.
Error correction is the real bottleneck in this field. Qubits are so fragile that large calculations are impossible without catching and fixing errors as they happen. IBM reports it has demonstrated real-time decoding of quantum errors on classical hardware, and that it did so a year ahead of schedule.
On timing there are two targets: the company expects the first verified cases of quantum advantage; where a quantum computer beats all classical systems; to be confirmed by the scientific community by the end of 2026. Fault-tolerant quantum computing is targeted for 2029. Both are IBM's own projections, and timelines in this field have slipped before.
Details are on the IBM Quantum platform.
