IBM has connected and cooled its first two quantum cryogenic modules together, an infrastructure designed to eventually link hundreds of chips within larger systems. Initial tests have brought the setup below 15 millikelvin, an engineering milestone that the company associates with its goal of building IBM Quantum Starling, its fault-tolerant quantum computer scheduled for 2029.
The key points of IBM’s new cryogenic systems in 30 seconds
- IBM has connected two cryogenic modules to operate within the same ultra-cold environment.
- The system first reaches 4 kelvin in less than five days and then drops below 15 millikelvin.
- Each module provides up to 12 times more wiring space than IBM’s most used quantum systems today.
- The architecture enables testing connections between multiple processors and is part of the pathway toward Starling in 2029.
While this announcement may seem less spectacular than unveiling a processor with more qubits, it addresses a fundamental physical challenge in scaling quantum computing: it’s not enough to just manufacture larger chips. They also need power, control, connections, and maintenance in extreme thermal conditions, without that infrastructure becoming a limiting factor.
A quantum computer needs more than just qubits
Superconducting quantum processors operate at extremely low temperatures. Under these conditions, the goal is to reduce thermal energy that can interfere with the delicate state of qubits.
That’s why large quantum systems use dilution refrigerators—multi-stage structures capable of approaching absolute zero, which is at 0 kelvin or -273.15°C.
IBM’s new facility advances this approach with a modular architecture.
The first two operational modules together exceed 2.4 meters in height and width. During initial tests, they reached 4 kelvin—around the temperature of liquid helium—in less than five days. Later, temperatures dropped below 15 millikelvin.
IBM compares this lowest temperature to deep space conditions and notes it is more than 180 times lower.
While this figure helps illustrate the operational environment, temperature isn’t the only critical aspect of this breakthrough.
Space also matters.
Each vacuum chamber offers up to 12 times more wiring space than IBM’s most used quantum systems. This additional capacity is designed to accommodate many more connections between processors, both within each module and across different modules.
This brings us to one of the main challenges for the next generation of quantum computers.
IBM aims to scale by connecting processors
The industry has been increasing qubit counts for years, but building monolithic processors of ever-greater size presents technical difficulties.
An alternative follows a path reminiscent, with all the differences inherent to quantum computing, of the evolution seen in other computing architectures: combining multiple modules to create larger systems.
IBM’s new cryogenic design resembles a box, allowing different units to be placed side-by-side in a single row.
Within this infrastructure, IBM uses so-called L-couplers, connections developed to link separate quantum chips and enable them to form larger systems.
The company plans to use this technology in 2027 to connect multiple processors and reach at least 1,000 programmable qubits. The roadmap envisions Nighthawk scaling up to nine modules of 120 qubits each, with up to 1,080 qubits and circuits of 10,000 gates over that period.
Next, they’ll need to evaluate how the infrastructure performs with actual processors.
IBM intends to install Quantum Nighthawk processors in the new cryogenic modules during 2026 to expand operational testing.
Currently, Nighthawk features 120 programmable qubits and uses a square grid topology where each qubit can connect to up to four neighbors. IBM considers it their platform for progressively increasing the complexity of quantum loads before moving to large-scale fault-tolerant systems.
The goal isn’t just to cram more qubits into a quantum fridge. IBM needs to demonstrate it can sustain cryogenic conditions and control multiple processors while dramatically increasing connection counts.
From Nighthawk to Starling: targeting 2029
A key piece to better understand this announcement is IBM Quantum Starling.
The company has long set 2029 as the year it expects to offer its first large-scale fault-tolerant quantum computer to clients.
Fault tolerance aims to overcome a major limitation of current quantum computing: qubits are highly sensitive to noise and can errors during operations. To run long and reliable algorithms, mechanisms are needed to detect and correct these errors mid-calculation.
IBM is developing an architecture based on error correction codes and multiple generations of processors and interconnections to achieve this.
Their roadmap includes Nighthawk for scaling circuit complexity, alongside architectures like Kookaburra and Cockatoo designed to demonstrate key components for future fault-tolerant systems.
Starling is the next major milestone.
IBM envisions the system will have 200 logical qubits and be capable of executing 100 million quantum operations or gates, significantly surpassing simply increasing physical qubit counts.
A logical qubit is constructed from multiple physical qubits plus error correction techniques. Therefore, the thousands of physical qubits housed in cryogenic modules should not be confused with the 200 logical qubits planned for Starling.
The company projects even further capabilities beyond that, with Blue Jay scheduled for 2033 or later, featuring up to 2,000 qubits and a capacity to perform 1 billion quantum operations.
These are business and technological goals, not current capabilities. IBM itself notes in its roadmap that these plans are its current intentions and may be modified.
Infrastructure is becoming as important as the processor
The new cryogenic system also signals an interesting evolution in quantum computing.
For years, much focus has been on processors and the number of qubits. As systems grow, the infrastructure surrounding the chip becomes more significant.
Refrigeration, control electronics, cabling, interconnects, error decoding, and high-performance classical computing systems must work in concert.
IBM Quantum System Two already embodies some of this philosophy. Its architecture combines scalable cryogenic infrastructure, modular control electronics, and classical servers, enabling multiple quantum processing units (QPUs) to operate within a data center environment.
Three essential components used in System Two have been incorporated into the new modular cryogenic architecture as well. The goal is to test and modify these components independently, rather than redesign the entire system each time one element changes.
The challenge will grow substantially in the coming years.
IBM acknowledges in its planning that reaching future systems like Blue Jay will require new control electronics and cryogenic infrastructure, as well as improvements to reduce space, energy consumption, and costs.
Therefore, connecting two large refrigerators and running them together below 15 millikelvin does not mean fault-tolerant quantum computing is achieved. It demonstrates a key engineering milestone necessary if IBM wants to transform hundreds of independent chips into a large-scale quantum machine.
The next challenge will be to do so with more processors, more connections, and increasingly complex error correction systems without sacrificing qubit stability.
Frequently Asked Questions
What temperature do IBM’s new cryogenic systems operate at?
Initial tests achieved bringing the connected modules below 15 millikelvin, after reaching 4 kelvin in less than five days.
Why does IBM need such large refrigerators for its quantum computers?
Superconducting processors require extremely low temperatures. Additionally, future systems will need much more space for wiring and connections among numerous quantum chips.
What is IBM Quantum Starling?
Starling is the fault-tolerant quantum computer IBM plans to make available to clients in 2029. The company projects it will have 200 logical qubits and the capacity to perform 100 million quantum operations.
Does this announcement mean IBM already has a fault-tolerant quantum computer?
No. The cryogenic system is a piece of the infrastructure needed to achieve that goal. IBM still needs to develop other advances in processors, interconnects, error correction, decoding, and software.
via: IBM Video
