LIVE

Nvidia, IBM, and US Laboratory Partner to Develop Hybrid Quantum Platform

⚖️ Regulation & Policy 2 min read 👁 7

Nvidia has teamed up with IBM and Fermilab to connect high-performance AI servers with quantum processors to tackle complex physics problems and quantum error correction.

Nvidia, IBM, and US Laboratory Partner to Develop Hybrid Quantum Platform⚖️Regulation & Policy

Photo: thenextweb.com

Key points

  • Nvidia, IBM, and Fermilab are building a joint platform linking AI supercomputers with quantum hardware.
  • The collaboration focuses on quantum error correction and scientific workloads in high-energy physics.
  • Nvidia released a decoder tool in its CUDA-Q library aimed at drastically shortening quantum simulation times.

Graphics hardware maker Nvidia is partnering with IBM and the US Department of Energy’s Fermi National Accelerator Laboratory to construct a hybrid quantum research platform. Announced early this week, the project integrates Nvidia DGX B200 systems alongside IBM quantum processors, featuring IBM's high-speed Nighthawk chip, to test how classical hardware and quantum systems can scale together for nuclear and high-energy physics applications.

Addressing Quantum Hardware Limitations

A key focus of the initiative is solving error-correction obstacles that currently restrict the scope and runtime of quantum computations. Fermilab will steer research into decoding algorithms and system requirements necessary for future fault-tolerant quantum architectures. In tandem with the announcement, Nvidia introduced a software decoder called TCBP-JOBS within its CUDA-Q QEC 0.9.0 library. The company claims the algorithm executes up to 13,000 times faster than current benchmark methods, reducing certain quantum system design workloads from nearly three months to under 45 minutes.

Neither partner disclosed financial terms or an official launch timeline for the project. Anna Grassellino, chief technology officer at Fermilab, referred to the initiative as a proposed platform, while IBM Fellow Jerry Chow noted that linking classical and quantum technologies creates new possibilities for scientific research. In a separate effort funded by Japan's NEDO agency, IBM and the RIKEN institute connected quantum hardware with Nvidia-powered supercomputers to model molecular interactions involving over 12,000 atoms for pharmaceutical applications.

To make hybrid systems easier to operate, Nvidia and IBM are also adding quantum job-scheduling tools directly into Slurm, an open-source management software widely used by scientific computing facilities. What to watch next: Researchers will track whether the partners establish concrete deployment dates and if Slurm's new features become the standard operational interface for hybrid supercomputing centers.

Source thenextweb.com — thenextweb.com/news/nvidia-ibm-fermilab-quantum-codesign-platform-decod… ↗

Contact the newsroom

Got a tip, a proposal or a partnership question? Write to us and we will reply shortly.