IBM has unveiled an ambitious plan to build the world’s first large-scale, fault-tolerant quantum computer by 2029. The announcement is supported by two new research papers and an updated quantum computing roadmap that sets a clear framework for achieving this groundbreaking technology. By 2029, IBM promises to deliver IBM Quantum Starling, a fault-tolerant quantum computer capable of running quantum circuits with 100 million quantum gates on 200 logical qubits.
The system is being developed at IBM’s historic facility, and the company is confident in meeting this goal based on its track record. Since 2020, IBM’s quantum roadmap has guided its efforts toward realizing practical quantum computing. The latest updates project a path extending to 2033 and beyond, with IBM having successfully delivered on each milestone to date.
IBM is currently the only organization positioned to run quantum programs at the scale of hundreds of logical qubits and millions of quantum gates by the end of the decade. IBM is a leader in quantum computing, with the only quantum computers capable of delivering accurate results for quantum circuits with over 5,000 two-qubit gates. Collaborations with partners such as RIKEN, Boeing, the Cleveland Clinic, and Oak Ridge National Laboratory have further bolstered IBM’s capabilities.
The company anticipates achieving quantum advantage by the end of 2026. However, current quantum devices are still limited to small circuits due to the limitations of error-mitigating techniques. Fully unlocking the potential of quantum computing will require devices capable of running larger circuits with hundreds of millions of gates and operating on hundreds of qubits.
These devices must also correct errors and prevent them from spreading.
IBM’s roadmap to fault-tolerant quantum computing
Quantum error correction involves encoding quantum information into physical qubits to protect against errors.
Similar to conventional computing, where redundant binary digits protect information, quantum error correction codes construct logical qubits from multiple physical qubits. This redundancy makes the information more resilient to errors. IBM has developed a modular, fault-tolerant architecture based on quantum low-density parity-check (qLDPC) codes, specifically the bivariate binary (BB) codes.
These codes demonstrate high error correction performance with significantly fewer qubits compared to traditional methods. For example, the [[144,12,12]] BB code encodes 12 logical qubits into 144 data qubits plus 144 syndrome check qubits, totaling 288 physical qubits. This scheme matches the performance of surface codes with only one-tenth the number of physical qubits.
IBM has also designed efficient, fault-tolerant logical processing units (LPUs) for these qLDPC codes. These LPUs perform logical measurements using low-weight checks and are highly scalable. Together with the symmetries of the qLDPC code, they facilitate essential quantum computations, including Clifford gates, state preparations, and measurements.
IBM’s fault-tolerant modular architecture, based on years of research and innovation, supports a scalable system for realizing fault-tolerant quantum computing. By following their roadmap and leveraging new technologies, such as the bivariate bicycle codes, IBM is poised to make strides toward practical, large-scale quantum computing in the near future. This announcement marks a pivotal step in IBM’s ongoing efforts to push the boundaries of quantum computing and bring its transformative potential closer to reality.
Senior Software Engineer with a passion for building practical, user-centric applications. He specializes in full-stack development with a strong focus on crafting elegant, performant interfaces and scalable backend solutions. With experience leading teams and delivering robust, end-to-end products, he thrives on solving complex problems through clean and efficient code.
