HPC
Quantum, integrated into your HPC center
Room-temperature neutral-atom QPUs that integrate into your HPC center, for hybrid quantum-classical workflows.
140+
reconfigurable qubits
20–21°C
Operating temperature
3
QPUs already deployed
in HPC centers
3.5kW
Average power consumption
FULL STACK
integrated
DEPLOYED
Give your users access to quantum computing within their HPC workflows.
Teams can explore quantum methods for simulation and optimization, develop and test quantum programs on emulators, and combine QPU execution with classical preparation and analysis. Start through cloud access and choose the level of integration that fits your center’s needs.
Published use cases
QUANTUM SIMULATION
One-to-one quantum simulation of the frustrated magnet TmMgGaO₄ with a 256-qubit Rydberg simulator, benchmarking simulated properties against experimental measurements and accessing non-equilibrium dynamics out of reach for classical resources.
INDUSTRIAL OPTIMIZATION
Transferable annealing protocols for Maximum Independent Set on neutral-atom QPUs, applied to EV smart-charging optimization on a real dataset, with experimental validation scaling to problems up to 100 qubits.
TELECOMMUNICATION
Hybrid quantum-classical Branch-and-Price method for Vertex Coloring, relevant to frequency assignment and network planning; includes experiments on real QPUs and reports optimal solutions in ~98% of benchmark instances.
How quantum fits into your users’ work
Classical resources prepare inputs and analyse results; the QPU executes the selected quantum computation.
Teams move between stages as their work requires.
CLASSICal
DEVELOP
Write quantum programs with Pulser, QoolQit, CUDA-Q or Qiskit, in the environments users already know.
CLASSICAL
EMULATE
Test and validate programs on emulators, locally or through Pasqal Cloud.
QPU
EXECUTE
Submit jobs through the scheduler to run the quantum computation on a QPU.
CLASSICAL
ANALYSE
Process results on classical resources and refine the next iteration.
ITERATE
Results from one stage feed the next, and teams return to development as often as their work requires.
Get started today
CLOUD ACCESS
Connect to Pasqal QPUs and emulators through Pasqal Cloud from any cluster with outbound internet access. Sessions can stay open for iterative hybrid workloads, and reservation windows support predictable scheduling.
SCHEDULER INTEGRATION
Through QRMI, Pasqal QPUs appear as a resource in your workload manager. Users request CPUs, GPUs and QPUs in the same job script, while policy and accounting stay with the center. Works with cloud-hosted and on-premises QPUs.
ON-PREMISES DEPLOYMENT
A Pasqal QPU and quantum access node inside your data center, running the same software stack, with dedicated capacity under your control. Requires facility and operations integration: power, cooling, network policy and monitoring.
x3
Pasqal QPUs
Already running in supercomputing centers
Pasqal’s HPC own integration stack, built around QRMI and Warden, is being used at CINECA, for Pasqal’s on-premises Sol QPU. At CEA/TGCC and Jülich Supercomputing Center, integration is provided through Bull’s Qaptiva solution. Pasqal supports both partner-led integration with Qaptiva and integration using its own software stack, according to each center’s requirements.
CEA
September 2025 – France
Pasqal Orion Beta — 100+ qubits
Jülich SC
November 2025 – Germany
Pasqal Orion Beta — 100+ qubits
Cineca – Leonardo
June 2026 – Italy
Pasqal Orion Gamma — 140+ qubits
An integration architecture built with the HPC community
Through scheduler integrations built around the open Quantum Resource Management Interface (QRMI), Pasqal QPUs can be allocated alongside classical computing resources. Users submit hybrid jobs through their HPC environment, while the resource manager retains control over allocation and site policies.
HPC workflows already combine specialized resources, and quantum computing is becoming one of them: a QPU executes selected quantum computations, while CPUs and GPUs keep running large simulations, pre-processing and post-processing. Integration has to fit the way the center operates, including its software, scheduling policies, facilities, networking and operational processes. Pasqal designs its QPUs and software stack around that fit, so quantum extends HPC as one more resource, not a separate system to run alongside it.
DESIGN PRINCIPLES
Familiar HPC workflows
Users access quantum resources through familiar job submission tools and development environments.
Hybrid by design
CPUs, GPUs and QPUs co-allocated in a single job.
Open source
QRMI, Warden, Pulser and QoolQit are open to the community.
Vendor-neutral integration
QRMI separates resource integration from vendor-specific device interfaces.
Clear ownership
Each layer has a clear owner, from resource management to quantum operations.
Users develop hybrid applications using tools such as Pulser, QoolQit (for neutral-atom quantum workflows), NVIDIA CUDA-Q and the Qiskit-Pasqal provider, according to the supported programming model. Workflows combine classical processing with quantum computations and can be developed and tested on emulators before compatible programs are executed on hardware.
Pasqal Orion: QPU specs
Reconfigurable qubits
Nominal number: 140
Max number: 200
Max Rabi frequency
2 MHz
Max detuning
7.75 MHz
Effective Computation
Repetition Rate
0.25 Hz (for defect free registers)
Peak power
~10 kW (avg ~3.5 kW)
Power supply
1×380V 3P+N+E 32A
Networking
2× Ethernet RJ45
Operating temperature
20-21°C
Footprint and weight
~200 × 590 × 390 cm; ~4,000 kg
When HPC and Quantum Work Together
Hear the perspectives of CINECA, GENCI and IBM during Pasqal Thoughts 2026 on how quantum computing is becoming a specialized accelerator inside hybrid HPC workflows rather than a replacement for classical supercomputing
QUANTUM ENHANCES THE HPC STACK
QPUs are no longer limited to the lab and are now entering the operating environment of HPC centers and supercomputing workflows.
SUPERCOMPUTING 2026
Meet Pasqal at SC26 on booth #5232 to explore how neutral-atom quantum computing can complement and accelerate high-performance computing systems.
In the news
FAQ
How do you integrate a QPU into an HPC center?
Integration connects user workflows, HPC resource management, quantum operations and the physical infrastructure. The implementation depends on the center’s requirements and can use Pasqal’s software stack or a partner solution such as Bull’s Qaptiva.
Which HPC schedulers support Pasqal QPUs?
Scheduler integrations built around QRMI enable Pasqal QPUs to be accessed from HPC workloads. Integration work covers Slurm, PBS, Open Cluster Scheduler and Flux, with capabilities depending on the scheduler and deployment.
What is QRMI?
QRMI, the Quantum Resource Management Interface, is an open-source, vendor-neutral interface and library through which resource managers and applications interact with quantum resources. It originated from an IBM initiative and was developed with Pasqal, STFC Hartree Center and Rensselaer Polytechnic Institute.
Do Pasqal QPUs work with NVIDIA CUDA-Q and Qiskit?
Yes. Pasqal supports NVIDIA CUDA-Q and provides Qiskit access through the Qiskit-Pasqal provider, within their supported programming models. See the CUDA-Q Pasqal documentation for supported capabilities and examples.
Does a neutral-atom QPU need cryogenics?
No. Pasqal neutral-atom QPUs need no dilution refrigerator: the system runs in a standard room-temperature environment (20 to 21°C).
Which HPC centers use Pasqal QPUs?
Pasqal QPUs are integrated at CEA/TGCC in France, Jülich Supercomputing Center in Germany and CINECA in Italy. At CEA/TGCC and Jülich, integration uses Bull’s Qaptiva solution.
Can we start without an on-premises QPU?
Yes. Teams can develop and test workflows on emulators, locally or through Pasqal Cloud, and execute compatible programs on available quantum hardware. Scheduler integration can support cloud-hosted as well as on-premises QPUs. Explore Pasqal Cloud.
Bring quantum into your HPC center
Talk to our team about integrating a Pasqal QPU into your infrastructure and workflows.