QuEra, HPE Partner on On-Premise Fault-Tolerant Quantum Systems
QuEra Computing and HPE have announced a collaboration to integrate QuEra’s neutral-atom, fault-tolerant quantum systems with HPE Cray supercomputing platforms, enabling on-premise hybrid computing. QuEra specializes in building neutral-atom quantum processors, while HPE provides the Cray supercomputing foundation that the two companies intend to pair with quantum hardware. This partnership offers HPC centers, national laboratories, and enterprises a way to install fault-tolerant quantum systems directly in their own environments, rather than accessing them solely via the cloud.
What the Partnership Entails
This collaboration focuses on developing integrated testbeds, co-designing hybrid algorithms, ensuring software interoperability, and benchmarking system-level performance, all with the goal of deployment alongside the HPE Cray platform. The two companies have been working together since 2025. At HPE Discover Las Vegas 2026, HPE showcased a magic-state demonstration featuring QuEra, illustrating their hybrid quantum-classical direction.
The partnership introduces an on-premise option to QuEra’s fault-tolerant roadmap. QuEra states that its first fault-tolerant machine, Libra, is a planned system with over 256 logical qubits, described as megaquop-class. It is designed to execute approximately one million reliable logical operations. QuEra aims for Libra to be available via the cloud in 2028, with a larger gigaquop system to follow. The on-premise track includes plans for a potential gigaquop system built with HPE.
The Intersection of Fault Tolerance and HPC
Fault-tolerant quantum computing relies on classical infrastructure because error correction requires rapid data transfer and intensive classical processing alongside the quantum hardware. In a hybrid workflow, the supercomputer handles extensive classical computations, while the quantum system addresses specific subproblems that classical methods cannot solve. This pairing explains why quantum software is increasingly designed for HPC-class execution. For instance, Fujitsu open-sourced its OpenQARP package, which includes an NVIDIA CUDA-Q backend and a 40-qubit simulator that runs across 1,024 PRIMEHPC FX700 nodes.
Yuval Boger, QuEra’s Chief Commercial Officer, noted that integrating neutral-atom fault-tolerant systems into an HPC environment presents a distinct engineering challenge compared to delivering quantum computing via the cloud. Masoud Mohseni, Director of HPE Quantum, stated that the partnership aims to create a pathway from current supercomputers to hybrid quantum-classical systems. Both companies view cloud access and on-premise deployment as complementary, allowing customers to develop applications in the cloud initially and then transition to an installed system later.