Industry·
NVIDIA Extends Slurm to Unify Quantum Processors in Supercomputing Centers
Slurm 26.11 adds a vendor-neutral orchestration layer so HPC sites can schedule QPUs alongside GPUs without custom integrations. Early adopters include IBM, IQM, Pasqal, Quantinuum, and Alice & Bob.

Quantum processor deployments at supercomputing centers have surged nearly sevenfold in five years, with roughly 30 sites now installing more than 50 QPUs. Because most of these centers already rely on the Slurm workload manager for traditional HPC jobs, NVIDIA’s community-driven extension lets them treat quantum hardware as just another schedulable resource — no bespoke plumbing required.
One Interface, Many QPUs
The new release builds on the established QRMI and QDMI standards, allowing each QPU vendor to publish a single configuration file that works across every Slurm-backed center. Alice & Bob, IBM, IQM, Pasqal, Quantinuum, and the MQSC consortium have already validated the approach, cutting integration effort from months to days.
For researchers, the payoff is immediate: hybrid quantum-classical applications written with NVIDIA CUDA-Q can now request GPUs and QPUs in the same job script, with Slurm handling placement, queuing, and accounting automatically.
Why it matters for GPU / AI infrastructure
As AI workloads increasingly explore quantum-accelerated kernels for optimization, chemistry, and error-correction decoding, a unified scheduler eliminates the operational friction of managing separate clusters. AiGpu customers running CUDA-Q on our GPU cloud can now prototype hybrid algorithms locally and deploy them to any Slurm-enabled quantum site without rewriting orchestration logic.
Complementary CUDA-Q updates — including a 20,000× faster TCBP-JOBS decoder, fault-tolerant compilation primitives, and GPU-accelerated pulse shaping — further shrink the feedback loop between algorithm design and hardware execution, a critical advantage for teams iterating on quantum-enhanced AI models.
- aigpu
- ai gpu
- ai gpu cloud
- aigpu dubai
- slurm
- quantum computing
- hpc
- cuda-q
By AiGpu Editorial · Editorial rewrite based on public reporting (NVIDIA Blog)
← All articles