Oak Ridge, IBM and AMD launch a shared platform to test quantum error correction

Started by Ellie40, Nov 08, 2026, 05:17 PM

Previous topic - Next topic

0 Members and 1 Guest are viewing this topic.

Topic: Oak Ridge, IBM and AMD launch a shared platform to test quantum error correction   Views(Read 52 times)
Active members in this topic:
Ellie40(1)

Ellie40

Three big names have teamed up to tackle the hardest problem in quantum computing. The Quantum Insider reports that IBM, AMD and Oak Ridge National Laboratory's Quantum Science Center have launched a co-design platform for testing quantum error correction alongside supercomputers

The first setup places IBM's 120 qubit Nighthawk processor next to Umbra, an Oak Ridge supercomputer built on AMD Instinct MI355X GPUs and fifth generation EPYC CPUs. It will also connect to the Oak Ridge Leadership Computing Facility, so classical and quantum machines can work on the same problems together

Error correction needs fast classical computing to decode errors as they happen, which is why pairing the two makes sense. Travis Humble, who directs the Quantum Science Center, said the aim is to build the foundation for more reliable quantum computers. Jay Gambetta of IBM Research went further, saying the question is no longer whether error correction will work on quantum hardware

The scale of the challenge is clear from the numbers. The US Department of Energy's Quantum Genesis Mission targets 100 logical qubits running 100,000 hard operations by late 2028, while IBM's own preprint shows up to 70 logical qubits and 438 operations so far. That leaves a gap of roughly 230 times to close in about two years

This is a first deployment with no results yet, so it is more of a starting line than a finish. Still, putting real quantum hardware inside a national lab's computing setup is exactly how the field moves from demos to dependable machines. Lessons learned here could feed straight into IBM's planned Starling fault tolerant computer

Can the gap to the 2028 target be closed in time? And will hybrid quantum and supercomputer setups become the standard way forward?