The revolutionary Starling system will feature 200 logical qubits capable of performing 100 million quantum operations with high accuracy, requiring approximately 10,000 physical qubits to create its logical qubits.12 This computational power is so vast that representing Starling's computational state would require the memory of more than a quindecillion (10^48) of today's most powerful supercomputers.23

The quantum computer will utilize IBM's proprietary quantum low-density parity check (qLDPC) codes, specifically their bivariate bicycle (BB) codes, which encode 12 logical qubits into 144 data qubits plus 144 syndrome check qubits.45 This approach achieves a 10x reduction in qubit requirements compared to surface codes while maintaining equivalent error correction capability, allowing the system to execute complex quantum operations with unprecedented stability and accuracy.67

The cornerstone of Starling's revolutionary capabilities lies in IBM's breakthrough quantum error correction technology. Using quantum low-density parity check (qLDPC) error correction code, IBM has achieved a remarkable 90% reduction in the number of physical qubits needed compared to traditional methods used by competitors like Google.12 This proprietary approach enables efficient and scalable error correction that addresses the fundamental challenge that has held back practical quantum computing applications.3

Real-time error correction will be implemented through classical hardware decoders that can read error syndromes and output corrected information.1 The system's gross code encodes 12 logical qubits into 144 data qubits, along with another 144 syndrome check qubits, for a total of 288 physical qubits while maintaining robust error correction capability.4 This technological advancement represents the culmination of years of research, with IBM stating that they have "de-risked the problems" faced in achieving fault-tolerant quantum computing, making them "more and more certain" of their ability to deliver on schedule.1

A modular architecture forms the foundation of Starling's design, enabling unprecedented scalability for quantum computing systems. The machine will consist of a network of interconnected modules, each containing quantum processor units (QPUs), all housed within the new Poughkeepsie data center facility.12 This innovative approach allows for increased qubit counts and computational capacity while maintaining system reliability.

The quantum data center will feature four IBM Quantum System Twos, each with a distinctive hexagonal layout accommodating three quantum processor units, complemented by conventional computing racks.23 Construction of this specialized facility began in April 2025, creating the infrastructure necessary to support the world's first large-scale fault-tolerant quantum computer.14 The modular design represents a significant advancement in quantum computing architecture, allowing IBM to scale beyond Starling to even more powerful systems like the planned IBM Quantum Blue Jay by 2033.53

The path to Starling follows a carefully planned sequence of increasingly advanced quantum processors, all named after birds. This progression begins with Nighthawk in 2025, the successor to the current Heron chip, which will eventually reach 15,000 gates by 2028.12 Following this, IBM will deploy the experimental Loon processor in 2025 to test qLDPC code implementation components.31

The roadmap continues with Kookaburra in 2026, the first modular processor combining quantum memory with logic operations, followed by Cockatoo in 2027, which will demonstrate entanglement between two Kookaburra modules using innovative "L-couplers" to link quantum chips like nodes in a larger system.12 This step-by-step approach culminates with Starling in 2028-2029, setting the foundation for an even more powerful system called IBM Quantum Blue Jay by 2033, which aims to execute 1 billion quantum operations over 2,000 logical qubits.4