IBM Connects First Modular Cryogenic Systems in Quantum Computing Milestone

IBM announced in August 2026 that it successfully connected two modular cryogenic systems, a milestone in the company's effort to build large-scale, fault-tolerant quantum computers.

The new architecture replaces traditional isolated cylindrical cryostats with a scalable, box-shaped modular design. By linking these cryogenic cells, IBM can create a shared ultra-cold environment where multiple quantum processors operate in tandem. The two connected modules cooled to below 15 millikelvin, the temperature required for superconducting qubits to function, in under five days.

Each modular cell provides about 0.53 square meters of wiring area and 2.75 cubic meters of vacuum chamber volume. The design offers up to 12 times more wiring space than IBM's previous widely used systems, which is critical for connecting hundreds of quantum chips.

The architecture uses L-couplers, cables capable of operating inside dilution refrigerators, to physically link separate quantum chips and allow information to move between processors. This lets multiple chips function as a single, more powerful quantum computer.

IBM plans to install its Quantum Nighthawk processors into these modules later in 2026 to begin operational performance testing. By 2027, the company intends to use the L-coupler technology to link multiple processors into a system with at least 1,000 programmable qubits.

The modular cryogenic architecture is a prerequisite for IBM Quantum Starling, the company's planned first fault-tolerant quantum computer, scheduled for delivery in 2029. IBM expects each cryogenic module to eventually house thousands of physical qubits, enabling large-scale, error-corrected quantum circuits.

Quantum computing researchers said the modular approach is significant because it allows each component to be independently tested and improved, reducing the risk of a single failure affecting the entire system.