IBM quantum computing cryogenic scaling

IBM is developing advanced cryogenic cooling technologies to support the scalability and stability of fault-tolerant quantum computers. Because superconducting qubits are highly sensitive to noise and heat, they must be maintained at temperatures below 15 millikelvin—a state described as “roughly 180 times colder than deep space.”

To achieve these temperatures, IBM utilizes helium compressors, dilution refrigerators, and multiple layers of thermal and electromagnetic shielding. The cooling process is a lengthy operation, often taking over four days to reach 4 Kelvin before descending to the required millikelvin threshold.

To address the challenges of managing heat in larger systems, IBM has demonstrated a modular architecture. This approach involves joining and cooling modular cryogenic systems into a single operating environment. Unlike traditional cylindrical designs, these new box-shaped aluminum modules can be placed side-by-side, providing up to 12 times more wiring space for high-density connections. This modularity is a critical step toward the development of IBM Quantum Starling, a system intended to overcome the physical limitations of single-chip scaling by allowing hundreds of quantum chips to operate within a shared, ultra-cold environment.

In a recent milestone at its Yorktown Heights facility, IBM successfully connected and cooled two modular cryogenic units, reaching temperatures below 15 millikelvin. This test utilized ‘L-coupler’ technology to support the roadmap for the IBM Quantum Starling, which is expected by 2029. Through these modular units, IBM aims to connect multiple processors to reach at least 1,000 programmable qubits by 2027. Alongside these hardware advancements, IBM introduced Qiskit Fermions, an open-source software framework designed to help researchers simulate fermionic systems on quantum processors.