IBM Runs Quantum Computing Algorithm on AMD FPGA Chips - TechPowerUp

Friday, October 24th 2025 IBM Runs Quantum Computing Algorithm on AMD FPGA Chips by Nomad76 Oct 24th, 2025 20:02 Discuss (4 Comments) IBM says it has successfully executed one of its key quantum computing algorithms in real time on AMD field-programmable gate arrays (FPGAs), marking a step toward cheaper and more practical hybrid quantum systems. The algorithm, designed to dynamically detect and correct quantum errors, was first announced in June and is now running on widely available AMD reconfigurable hardware rather than costly, custom-built control units. Jay Gambetta, IBM's VP of Quantum, told Reuters that the FPGA-based implementation performs 10× faster than required for live error correction, calling it "a big deal" for real-world quantum computing. The demonstration suggests IBM's development timeline is a full year ahead of schedule for its 2029 Starling quantum system roadmap. Quantum computers use qubits that can show both 0 and 1 at the same time but often face problems from outside factors that lead to errors in calculations. IBM's new method tackles this issue by keeping an eye on and fixing these errors as they happen, which keeps the system steady during quantum tasks. By running this quantum computing algorithm on regular AMD chips, we see how classical and quantum systems are coming together more and more opening doors to mixed setups where normal hardware handles fixing errors and running things in real time. IBM's approach might help close the gap between the early quantum machines we have now and future ones that don't make mistakes. By showing that off-the-shelf chips you'd find in data centers can do this job well, IBM cuts down on costs and makes things simpler while also making it easier to improve performance as systems get bigger. These are big steps in moving quantum computing out of labs and into the business world. The partnership between IBM and AMD appears to be bearing fruit. Back in August, the two companies announced a close collaboration to develop next-generation computing architectures that combine quantum computing and high-performance computing, a concept known as quantum-centric supercomputing. More recently, in early October, IBM and AMD also revealed a joint effort to deliver advanced AI infrastructure to Zyphra, an open-source AI research and product company based in San Francisco. Under a multi-year agreement, IBM will provide Zyphra with a large cluster of AMD Instinct MI300X GPUs hosted on IBM Cloud, enabling the training of next-generation multimodal foundation models.